You are on page 1of 132

Supported by:

Climate Change Scenarios –


Implications for Strategic
Asset Allocation
Public Report
Mercer wishes to thank the following organisations and individuals for their contributions to the
project. This research would not have been possible without the support and participation of the project
participants, Mercer’s team, Grantham LSE/Vivid Economics, the research group and a few other key
individuals.

Project participants:

International Finance Corporation (IFC)


Carbon Trust, United Kingdom
All Pensions Group (APG), Netherlands
Första AP-fonden (AP1), Sweden
AustralianSuper fund, Australia
British Columbia Investment Management Corporation (bcIMC), Canada
British Telecom Pension Scheme (BTPS), United Kingdom
California Public Employees’ Retirement System (CalPERS), USA
California State Teachers’ Retirement System (CalSTRS), USA
Environment Agency Pension Scheme, United Kingdom
Government of Singapore Investment Corporation (GIC), Singapore
Maryland State Retirement Agency, USA
Norwegian Government Pension Fund, Norway
Ontario Municipal Employees Retirement System (OMERS), Canada
VicSuper Pty Ltd., Australia
PGGM Investments, Netherlands

Mercer team:

Dr. Danyelle Guyatt (Primary Researcher, Project Manager)


Rob Curtis (Lead, Modelling)
Jianchun Wu (Modelling)
Harry Liem (Adviser, Modelling)
Susanna Jacobson (Researcher)
Dr. Jelle Beenen (Adviser)
Divyesh Hindocha (Peer Review, Adviser)
Crispin Lace (Adviser)
Nick Sykes (Adviser)
Rich Nuzum (Adviser)
Jane Ambachtsheer (Project design, Adviser)
Lynn Slipp (Project Management Support)
Laureen Bird (Project Management Support)
Kelly Gauthier (Analyst, Communications)
Matt Damsker (Editor)
Shaun Harding (Legal)
Helga Birgden (Peer Review, Client Management, AsiaPacific)
Craig Metrick (Client Management, US)
Dr. Elisabeth Bourqui (Client Management, Canada)
Rachel Whittaker (Researcher)
Vanessa Hodge (Researcher)
Dr. Xinting Jia (Researcher)
Rebecca Dixon (Researcher)
Deb Clarke (Specialist Input, Equities)
Amarik Ubhi (Specialist Input, Infrastructure)
Paul Cavalier (Specialist Input, Fixed Income)
Sanjay Mistry (Specialist Input, Private Equity)
Paul Richards (Specialist Input, Real Estate)
John Wills (Specialist Input, Real Estate)
The Grantham Research Institute on Climate Change and the Environment, London School of Economics
together with Vivid Economics:

Dr. Simon Dietz (Grantham LSE and Vivid Economics)


Dr. Sam Fankhauser (Grantham LSE and Vivid Economics)
Dr. Cameron Hepburn (Grantham LSE and Vivid Economics)
Dr. Alex Bowen (Grantham LSE)
Robin Smale (Vivid Economics)
Helen Jackson (Vivid Economics)
Philip Gradwell (Vivid Economics)
Dr. Robert Ritz (Vivid Economics)

Research group:

Alan Miller (Principal Climate Change Specialist Environment Department, IFC)


Dr. Monica Araya (Senior Associate, E3G, Third Generation Environmentalism, UK)
Ingrid Holmes (Programme Leader Low Carbon Finance, E3G, Third Generation Environmentalism, UK)
Professor Gordon Clark (Halford Mackinder Professor of Geography, Oxford University)
Nick Robins (Head of Climate Change Centre of Excellence, HSBC)
Joaquim de Lima (Global Head of Equity Quantitative Research, HSBC)
Bruce Duguid (Head of Investor Relations, The Carbon Trust)
Garrie Lette (Chief Investment Officer, Catholic Super, Australia)
Stephanie Pfeifer (Executive Director, Institutional Investor’s Group on Climate Change)

Other contributors:

Keith Ambachtsheer (Director, Rotman International Centre for Pensions Management, University of Toronto,
Canada)
Professor Rob Bauer (Director, European Centre for Corporate Engagement, Maastricht University, the
Netherlands)
Quotes from the partners on why they
participated in the research

“That climate change poses significant


financial and economic risks has only been
accentuated by the tens of billions of dollars
in losses due to recent climate-related natural
disasters such as the floods in Australia and
Pakistan and the wildfires in Russia. This
study makes a significant contribution to our
ability to measure the level of risk that climate
change creates for investment portfolios.
Managing that risk in a way that maintains
the returns expected by beneficiaries is a
crucial responsibility for the management
of these investment portfolios. This report
provides some practical steps that investors
can take today to shift their asset allocation
to manage climate change risks and finance
the much-needed infrastructure for a lower
carbon future.”
– Rachel Kyte, Vice President, IFC

“This report is unique and groundbreaking


in quantifying the increased portfolio
risk arising from global efforts to tackle Important notices
climate change. It demonstrates that Copyright 2011 Mercer LLC, Carbon Trust and
International Finance Corporation. All rights
unless this risk is tackled intelligently by reserved.

increasing exposure to climate-sensitive This report may not be modified or otherwise


asset classes, then long-term rewards could provided, in whole or in part, to any other person
or entity, without Mercer’s written permission. This
fall. The findings undermine the notion report may not be sold on a commercial basis.

of a conflict between ‘green’ investing The findings and opinions expressed in this
and acting in beneficiaries long-term report represent the intellectual property of the
authors as at the date of its publication. They are
financial interests. This will have profound not intended to convey any guarantees as to the
future performance of any investment products,
implications for fiduciary duties and places asset classes or capital markets covered by this
a clear obligation to increase analysis of report. Past performance does not guarantee
future results.
the consequences of climate change for
portfolio management.” This report does not contain investment advice.
As such, no investment decisions should be made
based on the contents of this report without first
– Bruce Duguid, Head of Investor obtaining appropriate professional advice from
Engagement, The Carbon Trust an independent financial adviser and considering
your own circumstances.

Information contained in this report has been


obtained from a range of third-party sources.
While the information is believed to be reliable,
no representations or warranties are given by the
authors as to the accuracy of the information
presented and no responsibility or liability
(including for indirect, consequential or incidental
damages) is accepted for any error, omission or
inaccuracy contained within the report.

Climate Change Scenarios – Implications for Strategic Asset Allocation


Contents
Executive summary 1 Methodology 93
n Collaboration 95
Report highlights 3 n Expanding the asset-allocation toolkit 97
n Scenario analysis 98
Overview of investment impacts 23
n Factor risk framework 99
n Source of investment risk across the climate 24
n Climate change risks – TIP™ framework 100
scenarios
n Sensitivity of assets to investment risks 25 Estimating the costs of adaptation 109
n Impact on equity risk premium 26
n Estimates of the TIP™ factor risks 28 Estimating residual damages 111
n Sensitivity of assets to the TIP™ risk factors 33
n Sensitivity of regions to the TIP™ risk factors 34 Table listing 115
n Quantitative analysis results 36
Glossary 117
Asset-class impacts 41
Bibliography 121
n Listed equities 43
n Fixed income 48
n Private equity 51
n Infrastructure 54
n Real estate 58
n Commodities 63

Country and regional impacts 69

Mapping evidence to the scenarios 75


n Uncertainties around the outcomes 76
n Macroeconomic impacts 77
n Technology investment 82
n Sector impacts 84
n Climate policy 86
n Physical impacts 89
“Why does climate change matter to institutional
investors like the Environment Agency pension fund? It
matters because we know that we will need to be paying
out pensions to our fund members well into the 21st
century. We think all pension funds will need to adopt a
climate change-proofed financial investment strategy in
the future to enable them to fulfill their fiduciary duties.
We also want our pensioners to retire into a similar
environment that we enjoy today and not one that is
affected by the extremes of climate change that could
reduce their life expectancy.”
– Howard Pearce, Head of Environmental Finance and
Pension Fund Management, Environment Agency

“In early 2010, we set a goal to better understand how climate change
could be factored into our broad investment actions. For example,
should the risk and return impacts of global warming modify our
allocation between and within asset classes? The Mercer study has
helped clarify our thinking on some of these uncertainties. In our view,
the report makes an original contribution by giving financial meaning
to recognised climate science (Stern, IPCC) and provides ideas on
constructing portfolios acknowledging climate trends. It also raises
many more questions and hopefully will stimulate additional in-depth
work around investment capital and climate change.”
– Doug Pearce, CEO/CIO, British Columbia Investment Management
Corporation (bcIMC)

“CalPERS has been a leading advocate for environmental


and climate change issues for many years and
recognises these to be key risks for long-term investors.
This opportunity to collaborate with institutional
investors from around the world to look at the impact
of climate change scenarios on investments helps us
to shape our strategic thinking in this area and better
integrate our programs, policies and risk management.”
– Joe Dear, CIO, CalPERS

Climate Change Scenarios – Implications for Strategic Asset Allocation


“Climate change is a global risk factor that all long-term
investors should take into account when formulating
investment strategy. This in-depth analysis will provide
valuable input to our long-term strategy reviews.”
– Tom A. Fearnley, Investment Director, Norwegian Ministry of
Finance, Asset Management Department

“Participating in this project has not only given us better


insight of what impact climate change could have on asset
classes and the long-term performance of our portfolio; it has
also given us enhanced tools for our strategic asset-allocation
analysis.”
– Johan Magnusson, Managing Director, Första AP-fonden (AP1)

“VicSuper has taken an active position in integrating


sustainability into its investment strategy. This has
involved investing in low-carbon equity funds such as the
Vanguard Carbon Aware International Shares Fund, as
well as in venture capital clean technology, which in turn
invests in technology and products providing solutions to
environmental challenges. Our participation in this Climate
Change Scenarios report has assisted our thinking in how
to integrate climate change risk and opportunity into our
investment strategy, and also in ways to access a robust
and defensible methodology to assess the possible risk and
return implications of climate change. We do this for the
benefit of our more than 250,000 members.”
– Peter Lunt, Head of Investment Research, VicSuper

“This project has given us insight into the


complexity of the effects climate change
could have on the risk and return of our
portfolio. Climate change proves to be a
source of uncertainty. Although there is
currently no straightforward answer to
managing this uncertainty, we will continue
to address this issue in our investment
activities.”
– Jaap van Dam, Managing Director Strategy,
PGGM Investments
Executive summary

It is widely acknowledged that climate change to absorb is estimated to amount to as much as


will have a broad-ranging impact on economies approximately $8 trillion cumulatively, by 2030.
and financial markets over the coming decades. Additional investment in technology is estimated to
This report analyses the extent of that impact on increase portfolio risk for a representative portfolio
institutional investment portfolios and identifies a by about 1%, although global investment could
series of pragmatic steps for institutional investors accumulate to $4 trillion by 2030, which is expected
to consider, including allocation to climate-sensitive to be beneficial for many institutional portfolios.
assets and the adoption of an “early warning” risk The economic model used in this study excludes
management process. physical risks of climate change which are not
consistently predicted by the range of scientific
n Traditional approaches to modelling strategic models, and primarily for this reason concludes
asset allocation fail to take account of climate that, over the next 20 years, the physical impact
change risk: Strategic asset allocation (SAA) is a key of changes to the climate are not likely to affect
component of the portfolio management process, portfolio risk significantly. However, this does not
with some research estimating that more than imply the absence of significant (and growing) risk,
90% of the variation in portfolio returns over time as shown by recent climate-related disasters that
is attributable to SAA decisions. While standard investors need to monitor closely. See Figure 1 for
approaches to SAA rely heavily on historical the contribution to risk for a representative portfolio
quantitative analysis, much of the investment mix.
risk around climate change requires the addition
of qualitative, forward-looking inputs. Given the
Figure 1
unclear climate policy environment and uncertainty
Contribution to risk for representative portfolio mix in ‘default’
around the full economic consequences of climate
case
change, historic precedent is not an effective
indicator of future performance. Equity risk premium
10%
1% Credit risk premium
n New approaches to Strategic Asset Allocation 5% Illiquidity premium
are therefore required to tackle fundamental Technology
shifts in the global economy: This report uses Policy
scenario analysis to anticipate future trends and 12%
develops four alternative pathways that might
result from climate change. Using the scenarios,
the report models climate change risks using 72%
the “TIP™ Framework”. This framework assesses
three variables for climate change risk: the rate of
development and opportunities for investment into Source: Mercer
low carbon technologies (Technology), the extent
to which changes to the physical environment will
affect investments (Impacts) and the implied cost n To manage climate change risks, institutional
of carbon and emissions levels resulting from global investors need to think about diversification across
policy developments (Policy). sources of risk rather than across traditional
asset classes: Mitigating climate change risks
n The “TIP™” framework suggests that climate will require a new approach for investors. The
policy could contribute as much as 10% to overall short-term horizon of traditional equity and bond
portfolio risk: Uncertainty around climate policy is investments means that it will be more difficult for
a significant source of portfolio risk for institutional investors to price in long-term risks around climate
investors to manage over the next 20 years. The change compared to some of the more climate
economic cost of climate policy for the market sensitive assets. Consequently, the traditional way

1 Climate Change Scenarios – Implications for Strategic Asset Allocation


of managing risk through a shift in asset allocation well as help tackle the wider challenge of climate
into increased holdings of more conservative, change by increasing investment in mitigation and
lower risk, lower return asset classes may do little adaptation efforts globally. These results imply that
to offset climate risks. Further, in some scenarios typical funds are likely to require a shift in allocation
such a strategy could result in a decline in returns, towards more climate sensitive investments, as most
adversely affecting long-term portfolio performance will have only limited holdings in these classes. The
and potentially affecting income for beneficiaries. extent of any shift will also depend on the overall
view of the probability of different scenarios taking
n Managing climate change risks could lead to place.
increased allocation to climate sensitive assets:
This report finds that under some scenarios, the n Investors can take steps now to improve the
best way to manage the portfolio risk associated resilience of their portfolios to climate-related risks:
with climate change, while retaining similar returns, This report proposes a series of pragmatic steps
is to increase exposure to those assets that have a that investors can take today to begin the process
higher sensitivity to climate change “TIP™” factors. of managing climate change risks. Initial actions
The analysis suggests that under certain scenarios, could include the following: introduce a climate risk
a typical portfolio seeking a 7% return could manage assessment into ongoing strategic reviews; increase
the risk of climate change by ensuring around 40% asset allocation to climate-sensitive assets as a
of assets are held in climate-sensitive assets (this climate “hedge”; use sustainability themed indices
includes opportunities across a range of assets in passive portfolios; encourage fund managers
including infrastructure, real estate, private equity, to proactively consider and manage climate risks;
agriculture land, timberland and sustainable listed/ and engage with companies to request improved
unlisted assets) – see Figure 2 for an example disclosure on climate risks. It also highlights
of asset class portfolio mixes by scenario. Some the need for investors to communicate with
of these climate sensitive investments might policymakers the need for a clear, credible and
be traditionally deemed as more risky on a internationally coordinated policy response and for
standalone basis, but the report shows that selected dialogue to emphasise the potential economic and
investments in climate-sensitive assets, with an financial cost of delay. While many institutional
emphasis on those that can adapt to a low-carbon investors might view engagement with policymakers
environment, could actually reduce portfolio risk as a separate function from strategic decision-
in some scenarios. This offers the prospect that making processes, the findings of this study suggest
institutional investors’ interests can be aligned to that it can play a vital role in overall portfolio risk
both serve their beneficiaries’ financial interests as management.

Figure 2
Example of portfolio mix across the scenarios – portfolio to target 7% return

100% Cash
90%
Sovereign fixed income
80%
Credit (investment grade)
70%
Developed equity
60%
Emerging market equity
50%
Private equity (including
40%
renewables)
30% Sustainability/renewable
20% themed equities
10% Timberland/agriculture

0% Real estate (core)


Default Regional Delayed Stern Climate
(11% risk) Divergence Action Action Breakdown
(11% risk) (14% risk) (9% risk) (12% risk)

Source: Mercer

2
Report
Highlights
“Technologies change, competitive
structures change, government policies
change, and the way in which they operate
change. If we are going to have markets
that work well tomorrow, we must be
continually concerned that they are
going to adapt to new problems and new
strategies.”
– North (1999:24)

3 Climate Change Scenarios – Implications for Strategic Asset Allocation


Climate change is a systemic risk

Climate change was described by Nicholas Stern as


“the greatest market failure the world has seen” (Stern
Review, 2007). But relatively little research has focused
on the investment implications of climate change at
the total-portfolio level and how institutional investors
might respond. That is the purpose of this project.

Uncertainty is a key stumbling block in climate-


change research. Every link in the chain of manmade
greenhouse gas emissions, physical changes in the
climate system and their socioeconomic impacts is
highly uncertain. Therefore, investors cannot simply
rely on a best guess as to how the future will unfold
when planning their investments. Moreover, because
many of these uncertainties emanate from complex
systems that are poorly understood and difficult to
model, climate change has been called a problem of
“deep uncertainty” (Lempert, Groves et al, 2006).

In this context, deep uncertainty implies that


probabilities cannot be assigned to future states
with high confidence. This calls into question the
appropriateness of relying too heavily on quantitative
modelling tools, for which investors must specify
probability distributions to underpin the parameters of
their investment models.

Institutional investors must develop new tools to


more effectively model systemic risks such as climate
change. These tools require an expansion of the way
we think about portfolio risk, looking beyond mere
volatility. Describing probable scenarios, identifying
the potential sources of risks, and measuring and
monitoring them over time are the components of
an improved risk management strategy that seeks
to protect the long-term assets that institutional
investors oversee on behalf of their stakeholders.

It is in this context that the collaborative group came


together to look at the implications of climate change
for strategic asset allocation (SAA). Box 1 (on page
5) summarises the role of SAA in the institutional
investment management process. Led by Mercer, 14
global institutional investors, the IFC and the Carbon
Trust all joined forces to examine what climate change
might mean for the underlying drivers of the major
asset classes and regions around the world. Grantham
LSE/Vivid Economics and a research group composed
of specialist practitioners and academics were also
involved in parts of the process along the way.

4
Hightlights Box 1:
Systemic risk and the role of strategic asset allocation

SAA can be broadly defined as the use of optimisation international/emerging equity split, duration of fixed
tools by asset owners to determine long-term asset income and the split between nominal and inflation-
allocation benchmarks to achieve their long-term adjusted fixed income, allocation to unlisted assets and
objectives. The objectives vary depending on the type sustainability-themed assets. This is distinct from other
of asset owner and its obligations to beneficiaries or considerations such as portfolio structuring (including
other stakeholders. For example, the objective may be allocation to capital weightings, styles and sectors, and
to generate sufficient returns to hedge liabilities, to includes active/passive analysis) and manager selection
protect a reserve pool of assets while minimising risk and (the evaluation of manager performance in order to
maximising return, to minimise variations in contributions select one suitable for a client’s requirements).
for sponsors, or to target a certain funding level.
Below is a visual depiction of the distinction between
SAA involves making decisions about allocation to high- SAA decisions and other investment decisions.
level asset classes – that is, equity/fixed split, domestic/

Strategic asset
allocation decisions

n Equity/fixed income split


n Fixed income duration
n Domestic/foreign equity split
n Market risk/active risk split

Returns-based analysis Holdings analysis

n Risk/return tradeoffs n Value/growth vs. core


n Alpha n Large/mid/small
n Tracking error
n Net, gross of fees
n Active/Passive

Manager allocation

n Structure determined by both


returns
n Desired volatility can be refined at
the sub-asset class level
n Potential new managers can be
evaluated for fit

5 Climate Change Scenarios – Implications for Strategic Asset Allocation


SAA is a key component of the portfolio management
process, with academic research estimating that more
than 90% of the variation in portfolio returns over time
are attributable to SAA.1 When considered just in terms
of contribution to returns, SAA dominates over market
timing and security selection.

This backdrop was relevant for considering the


investment implications of climate change, as many
investors have, to date, approached climate change
primarily from a bottom-up, opportunistic perspective,
investing in climate-sensitive securities and assets when
opportunities arise. While this is important, it addresses
only part of the picture.

Additional consideration should be given to exploring


what climate change might mean for the underlying
determinants of asset-class risk and return, as well as for
overall market risk. Bottom-up analysis may not in itself
be sufficient to reveal market shortcomings in the pricing
of systemic risks ahead of time, which potentially leaves
institutional investors exposed to unexpected adjustment
costs from large-scale events, as the global financial crisis
has reminded us.

It is therefore prudent for institutional investors to


work towards building in, ahead of time (to the extent
possible), potentially large-scale systemic risks, such as
climate change, into risk management and SAA decision-
making processes.2 This requires the development of a
framework to unravel the uncertainties around climate
change, combining both top-down and bottom-up tools
and processes.

1 See Brinson et al (1986); Grinblatt and Titman (1989); Brinson et al (1991); Blake et al (1999); and Ibbotson and Kaplan (2000).
2 See ”Beyond the Credit Crisis: The Role of Pension Funds in Moving to a More Sustainable Capital Market” (2009), available at
http://www.mercer.com/referencecontent.htm?idContent=1332305.

6
Traditional asset allocation methodologies where diversification across assets is sought.
do not adequately capture climate change An additional tool for this analytic approach is
to think of SAA in terms of diversifying across
risks
sources of risk, rather than via asset classes per
se. This means utilising a factor risk approach to
Traditional modelling approaches do not adequately
supplement asset-allocation decision making.
capture the nature of the economic transformation
process and the potential source of risks associated
with climate change. As such, the tools to integrate 4. Need to be more forward looking: Climate
change requires forward-looking analysis and
climate change into the way we think about SAA
cannot rely on the traditional technique of
risk must be expanded to reflect the following:
modelling historical asset-class relationships.
This means utilising tools such as scenario
1. Need to embed climate change risk into
asset-allocation processes: Climate change
analysis.
can have a significant impact on the
performance of a portfolio mix over the 5. Need to go beyond quantitative analysis:
Qualitative factors need to be embedded into
long term, with the primary source of risk
the decision-making process. SAA decision-
resulting from uncertainty about climate
making processes rely heavily on quantitative
policy and its associated adjustment costs.
analysis, whereas much of the investment risk
The findings of this study show that for most
around climate change requires the exercise
asset classes, the impact of climate change
of judgement about how things might develop
varies significantly across different scenarios,
in terms of the science of climate change,
contributing as much as 10% to portfolio risk
the policymakers’ response and the types of
for a representative asset mix. This supports
technologies that may or may not prosper.
the need for a clear climate policy framework
as well as ongoing analysis to build these risks
into asset-allocation models. 6. Need to review assumptions regarding market
risk: Past periods of economic transformation
have been associated with a significant change
2. Need to look beyond macroeconomic impacts:
The Grantham LSE/Vivid Economics analysis
in the realised equity risk premium (ERP)3 over
time, ranging from destructive war-time periods
showed that the potential impact of climate
to positive periods of substantial efficiency
change on GDP, interest rates and inflation
improvements arising from a growing service
across the scenarios magnifies beyond 2050 but
sector and innovations in IT. Assumptions
will not be the driving force behind investment
regarding the ERP should therefore be reviewed in
risks before then. Mercer’s analysis indicated
light of the potential impacts of climate change
that the source of investment risk over the
on the process of economic transformation that
coming 20–30 years will result from increased
may occur in the transition to a low-carbon global
uncertainty about new technology, physical
economy.
impacts and climate policy (called the TIP™
factor risk framework).

3. Need to think about diversification across


sources of risk: To varying degrees, traditional
asset allocation techniques optimise portfolio
exposure based on assumptions about the risk,
return and correlation between asset classes

3 Broadly defined, the ERP represents the compensation for taking on equity risk versus a risk-free rate.

7 Climate Change Scenarios – Implications for Strategic Asset Allocation


A new framework has been developed to Our framework is built on three elements:
unravel climate change uncertainties
n Developing factors to represent the investment
Our goal in this project was to develop a framework to impacts of climate change and linking these factors
put around climate change that will assist institutional to the key drivers of different asset returns
investors in their risk management and SAA processes. n Developing climate-change scenarios and an
understanding of how climate change and asset
The study’s time horizon focused on the potential classes may respond in each hypothetical scenario
investment impacts out to 2030. The reason for this
n Building a simple quantitative framework to test the
is that while strategic investment decisions may be
relationships established in the factor analysis and
reviewed on an annual basis, they are typically set with
to decide whether any investor action is appropriate
a 10+ year horizon in mind. The time path of potential
impacts out to 2050 was also considered, to provide
To better analyse the investment impact of climate
investors with a sense of how things might evolve.
change, Mercer developed the TIP™ risk factor
framework (Figure 1) to examine which factors drive
The key questions to address are:
asset-class returns into the following three areas:

1. What investment risks and climate change


issues must institutional investors take into
n Technology (T) – broadly defined as the rate of
progress and investment flows into technology
account as part of their strategic decision-making
related to low carbon and efficiency, which are
processes?
expected to provide investment gains
Impacts (I) – the extent to which changes to the
2. What impact could climate change have on n

different asset classes and regions? physical environment will affect investments
(negatively)

3. What actions can institutional investors take? n Policy (P) – the cost of climate policy in terms of the
change in the cost of carbon and emissions levels
4. What are the messages for climate change
policymakers?
that result from policy, depending on the extent to
which it is coordinated, transparent and timely

These factors are interdependent; hence, the


framework cannot be viewed in a linear way.
Each factor is a key consideration in future asset
performance.

Highlights Figure 1
TIP™=Technology, Impacts and Policy
Factor risk approach to evaluate climate change investment impacts
Investment in energy efficiency, technology
development and deployment

Technology

Physical changes to our environment, Impacts Policy


Changes to carbon costs and emissions
health and food security (Physical) levels as a result of policy measures

Source: Mercer

8
Our goal was not to produce a quantitative analysis that necessitate relatively high levels of adjustment
leads to a statistically optimal portfolio for all investors. costs to comply with the new regulations. After
Indeed, given the uncertainties, we believe that such an the introduction of regulatory changes, the level of
aim is unrealistic. Instead, the framework is intended uncertainty regarding climate policy will decline,
to help investors gain additional insight into the risks creating a stronger investment backdrop.
within their current investment policies and decide
how best to try to manage the added risks arising from n Stern Action – This scenario has been named to
climate change. reflect the policy response advocated by Nicholas
Stern, author of the Stern Review (2007). It is the
In considering how climate change might have an most aggressive scenario in terms of policy response
impact on a portfolio’s asset mix from now until and private-sector innovation. It suggests that
2030, four scenarios were developed, the key features there will be swift agreement to a global framework
and outcomes of which are summarised below. The and a very high level of coordination in policy
scenarios do not represent a forecast of the future and efforts internationally, resulting in a high degree
should not be interpreted in a probabilistic way; rather, of economic transformation across the global
they provide a framework for considering the key economy, with new investment opportunities as
climate change drivers from an investment perspective well as risks. The uncertainties are lower than for
over the coming decades. A broad indication as to the other scenarios, as investors are able to predict
which scenario is more or less likely to have an impact the pathways of policies with a reasonable degree
is indicated in Table 1 (on page 10) to provide some of confidence, as policies are implemented in a very
general guidance for interpretation. The likelihood was transparent and orderly manner internationally. This
based on discussions among Mercer, Grantham LSE/ scenario will be associated with a higher economic
Vivid Economics and the Research Group. cost, in order to achieve the level of abatement in
emissions; however, the GDP impact is expected to be
n Regional Divergence – Some regions (EU and secondary in driving asset-class returns within our
China/East Asia) demonstrate strong leadership in report’s time horizon. Less uncertainty for investors
responding to the need to reduce emissions and about climate policy and new technology investments
act locally, with policy mechanisms ranging from will be the major drivers of positive transformation.
market-based to regulatory solutions. Other regions
(Russia) fail to respond and continue their high n Climate Breakdown – The status quo prevails in
levels of emissions. Some regions (US, India/South terms of policy, business and consumer behaviour.
Asia and Japan) fall somewhere in the middle, with With continued reliance on fossil fuels, carbon
local initiatives and measures associated with high emissions remain high and there is little economic
policy implementation risk. Overall, this scenario transformation. The investment impacts are hard to
involves a high degree of economic transformation predict, although the risk of catastrophic climate-
and investment in some regions, but the level related events increases significantly over time,
of uncertainty increases for investors due to the reaching critical levels towards the end of this
disparate nature of the policy responses across the century. This scenario brings potentially very high
different regions, increasing market volatility. risks for investors over the long term, particularly for
regions, assets and sectors that are most sensitive to
n Delayed Action – Business as usual (BAU) continues the physical impacts of climate change.
until the year 2020, when rapid policy measures
will be introduced that will lead to significant shifts
in behaviour that raise the cost of fossil fuel usage
dramatically (such as a global carbon tax) and quickly
reduce emissions. There is a high degree of economic
transformation led by public sector regulation
rather than by private sector innovation; this will

9 Climate Change Scenarios – Implications for Strategic Asset Allocation


Highlights Table 1
Key features and potential outcomes of the climate scenarios to 2030

Scenario Global policy response Carbon cost (in 2030) Emissions levels (now to 2030)

Regional Divergent and unpredictable Cost of carbon $110/tCO2e in 50 Gt4 CO2e emissions per year
Divergence – Framework agreed to all countries in this study (EU, in 2030 (equivalent to
succeed Kyoto Protocol US, China/East Asia and Japan) -20% from BAU)
(Most likely) – Targets announced of except India/South Asia and
medium ambition Russia

Delayed Action Late and led by hard policy Cost of carbon $15/tCO2e 40 Gt CO2e emissions per year
measures to 2020, then dramatic rise in 2030 (equivalent to
(Close second – Strong mitigation, but only to $220/tCO2e globally (not -40% from BAU)
in likelihood) after 2020, when sudden unanticipated by the market)
drive by major emitting
nations results in hasty
agreement
– Very little support to
vulnerable regions on
adaptation

Stern Action Strong, transparent and Cost of carbon $110/tCO2e 30 Gt CO2e emissions per year
internationally coordinated globally (anticipated by the in 2030 (equivalent to
(Much less action market) -50% from BAU)
likely) – Generous support to
vulnerable regions for
adaptation

Climate BAU; no mitigation beyond Cost of carbon $15/tCO2e 63 Gt CO2e emissions per year
Breakdown current efforts limited to the EU Emissions in 2030 (equivalent to BAU)
– Very little support to Trading Scheme regional
(Least likely) vulnerable regions for schemes and implicit cost of
adaptation carbon estimates
Source: Grantham Research Institute LSE/Vivid Economics

4 “Gt” refers to gigatonne, which equals 1,000 million tonnes of CO2e emissions.

10
Key findings of climate change impacts on
investments
2. Technology investments could accumulate to
$5 trillion by 2030: The private-sector response to
changing environmental conditions, new
1. Climate change increases investment risk:
Climate change increases the uncertainty
technology and policy measures may produce
a substantial number of new investment
and event risk that could have an impact on opportunities. According to Grantham LSE/Vivid
the realised returns for risky assets across Economics, by 2050 fossil-fuel use could decline
the scenarios, with higher risk resulting from by as much as two-thirds under Stern Action.
inefficient policy (see Table 2). Figure 2 shows the shift in energy demand and

Highlights Table 2
Impact of scenarios on source of investment risks

Scenario Fundamental Market


Climate change factors
factors factors

Economic cycle ERP Technology Impact Policy


Inflation Volatility

Regional Unchanged Higher High dispersion of Higher risk of future Higher uncertainty
Divergence volatility capital inflow into low- impact costs due to and potentially higher
carbon investments; slower reduction in reward for some
leading countries emissions assets due to regional
include the EU and disparity in climate
China policy

Delayed Higher Higher Business as usual Higher risk of future Higher uncertainty
Action inflation volatility (BAU) investment in impact costs due to around policy until
Higher interest Lower low carbon until 2020 delay in policy response 2020, then dramatic
rates realised ERP when policy measures U-turn reduces policy
stimulate flows uncertainty

Stern Unchanged Lower Clarity on climate policy Lower risk of future Policy clarity at the
Action volatility stimulates strong capital impact costs due to global level reduces
Higher flows into low-carbon reduction in emissions investment uncertainty
realised ERP solutions

Climate Unchanged Unchanged; Higher risk attached to Higher impact risks due BAU climate policy
Breakdown risk of low-carbon technology to lack of policy action, (unchanged from
higher investments due to rising future costs and today’s measures)
volatility policy inaction market pricing in future
policy shift
Source: Mercer

11 Climate Change Scenarios – Implications for Strategic Asset Allocation


supply under Stern Action. About two-thirds of
the shift is attributable to lower overall energy
3. Impact costs could accumulate to $4 trillion
by 2030: Grantham LSE/Vivid Economics have
demand, primarily due to improvements in energy
estimated that the cumulative economic cost of
efficiency, while the remaining third results from
changes to the physical environment, health and
supply-side changes. Mercer estimates, based on
food security across the climate scenarios could
International Energy Agency data, suggest that
be in the range of $2 trillion to $4 trillion by 2030,
additional cumulative investment in efficiency
with costs rising the greater the delay and the
improvements, renewable energy, biofuels, and
less well-coordinated the policy response. Most
nuclear and carbon capture and storage (CCS)
adaptation costs come from infrastructure (for
could expand in the range of $3 trillion to $5
example, transport and coastal zone protection,
trillion by 2030 across the mitigation scenarios
such as flood defence) sectors; though in Africa,
examined in this study. This presents meaningful
water supply and agriculture comprise more than
investment opportunities that are still in their
half of all costs (see Figure 3).
infant stages.

Perhaps the most important issue that is not


reflected in these estimates is the impact of
Highlights Figure 2
climate change in the longer run. Since many of
Renewables and nuclear overtake fossil fuels, in Stern Action
the greenhouse gases emitted today (particularly
scenario, by 2050
CO2) might still reside in the atmosphere until
Total energy demand Carbon capture storage (CCS) 2100 and beyond, emissions reductions are
required in the short term in order to avoid them.
Fossil without CCS Renewables + nuclear
As a result, consistent with the Stern Review
(2007), the cost of climate change will rise rapidly
600
after 2050.

500 490 It is also important to bear in mind that the direct,


462
446 economically realised costs of climate change
423
may reflect only a fraction of total costs incurred,
400
Energy supply (EJ)

particularly in developing countries. Property


351
379 insurance, for example, is much more extensive
319
300
in the industrialised world than it is in developing
countries, such that many losses in the latter
231
may be uncompensated but nevertheless real.
200 By way of illustration, costs incurred from the
128 Pakistani flood damage in 2010 were calculated to
123
be up to $43 billion. Climate damage is therefore
100 88
67 64 an important risk for institutional investors to
48
24 manage, both in terms of asset sensitivity and in
0 terms of influencing policy outcomes to mitigate,
0
2010 2020 2030 2040 2050
and adapt to, these risks.

Source: Grantham LSE/Vivid Economics, based on Edenhofer et al (2009)

12
Highlights Figure 3
Adaptation costs in 2030 for Climate Breakdown scenario
Infrastructure Agriculture

Coastal zone protection Fisheries

Industrial and municipal Human health


water supply and riverine
flood protection Extreme weather events

Sub-Saharan
Africa

India and
South Asia

MENA

Latin America
and Caribbean

Russia and the


former Soviet
Union
China and
East Asia

-5 0 5 10 15 20 25 30
$US billion
Source: Grantham Research Institute/Vivid Economics calculations, based on World Bank (2009a)

4. Policy measures could increase the cost of of approximately $15/tC02e. These costs may be
explicit in the market or implicit costs that
carbon emissions by as much as $8 trillion
affect operating costs outside of emission trading
cumulatively, by 2030: The future cost of carbon
schemes.5
emissions increases the longer the policy delay
and the less well-anticipated and coordinated the
policy action is. Grantham LSE/Vivid Economics
has estimated that the cost of carbon could
be $110/tC02e to $220/tC02e by 2030 across the
mitigation scenarios, compared to the current EU
Emissions Trading Scheme (ETS) price equivalent

5 For a discussion of the implicit price of carbon and estimates, see Vivid Economics, The Implicit Price of Carbon in the Electricity Sector of Six Major Economies,
October 2010, available at http://www.interactivemediarelease.com/ogilvy/ClimateInstitute.

13 Climate Change Scenarios – Implications for Strategic Asset Allocation


Highlights Figure 4
Climate change risks – TIP™ framework formulation

Technology: $ size of Impacts: $ cost of physical Policy: $ change in cost of


additional low carbon climate change impacts by emissions to 2030 as a result
investment flows by 2030 2030 of climate policy

Cumulative additional Cumulative economic cost Change in cost of emis-


investment in efficiency of changes to the physical sions = [2030 Emissions x $
improvements, renewable environment, health and food / tCO2e] – [2010 Emissions
energy, biofuels, nuclear and security to 2030 (Source: x $ / tCO2e] (Source: CAIT
CCS to 2030 (Source: derived estimates by Grantham LSE/ and Grantham LSE/Vivid
by Mercer from IEA WEO Vivid Economics) Economics)
2009)

IEA estimates modified Calculations by Grantham Carbon price derived


according to different LSE/Vivid Economics, using by Grantham LSE/Vivid
degree of mitigation across Hope’s PAGE2002 model Economics from the WITCH
scenarios. Climate Breakdown estimates and data on model; emissions derived
is baseline investment flows adaptation costs from the by Grantham LSE/Vivid
that would happen without World Bank/United Nations Economics based on Bowen &
additional mitigation Framework Convention on Ranger, 2009 and IEA 2009
Climate Change (UNFCCC)

Result: The value of additional Result: The costs range in Result: The increase in the
investments in these assets the order of $70 bn to $180 cost of emissions from 2010
will grow by between $180 bn pa globally in terms of to 2030 ranges between
bn to $260 bn pa to 2030 for adaptation and residual $130 bn and $400 bn pa
all mitigation scenarios, with damage costs, with Climate globally, with Delayed Action
Stern Action at the upper end Breakdown the highest cost the most costly due to late
and unanticipated policy

Source: Mercer. The factors have been discounted to the net present value using a 3% discount rate. This was chosen based on a composite of global 10-year bond
yields as at October 2010.

14
5. Infrastructure, private equity, real estate and
some commodities are highly sensitive to
6. Sustainable assets could act as a hedge: As
Figure 5 highlights, sustainable assets perform
climate change: The results of the asset-class comparatively well across the mitigation scenarios
impacts are summarised in Table 3, where the compared to core assets.6 The exception to this
overall sensitivity of each asset-class to the is Climate Breakdown, which is not surprising, as
climate-change TIP™ risk factors is presented in this assumes no further progress on policy from
the highlighted section at the top of the table, where we are today. Exposure to sustainable-
with the direction of the impact (positive, negative themed equities, efficiency/renewables in listed
or neutral) denoted by the colour. and unlisted assets, timberland and agricultural
land could therefore improve the resilience of a
portfolio mix across the climate scenarios.

Highlights Table 3
TIP™ factor risk sensitivity and direction of impact for asset classes
Sensitivity of the impact: where L = Low; M = Moderate; H = High; VH = Very high sensitivity to the combined climate change factors.

Direction of the impact: where = Positive; = Neutral; and = Negative. Agriculture = agricultural land; RE = real es
estate;
Infra = infrastructure; EME = emerging-market equity; EMD = emerging-market debt; LBO = leveraged buyout; VC = venture capital.

Listed equities Fixed income Commodities RE Private equity Infra


Efficiency/renewables

Efficiency/renewables

Efficiency/renewables
Sustainable equity

Agricultural land
Inv grade credit

Core, unlisted
Global equity

Global fixed

Timberland

Unlisted
EMD
EME

LBO

VC

Sensitivity L M H VH L M L H H H M H VH H VH

Regional Divergence

Delayed Action

Stern Action

Climate Breakdown
Source: Mercer. Sustainable equity = broad multi-themed listed equity companies that generate a substantial proportion (typically more than 25%) of their earnings
through sustainable activities. Efficiency/renewables assets = both listed/unlisted sustainability themed assets whose core activities are theme specific and more
concentrated in terms of exposure than are broad sustainability equity. This includes (but is not limited to) energy efficiency, low energy transport, renewable energy,
bioenergy, carbon capture and storage, smart grid, water supply, usage and management, waste management, hydro energy and geothermal, to name a few.

6 “Sustainable assets” refer to investments that generate a substantial proportion (typically, more than 25%) of their earnings through sustainable activities. At its
broadest level, sustainable investment seeks to support sustainable economic development, enhance quality of life and safeguard the environment. This includes
sustainable themes such as energy efficiency, low energy transport, renewable energy, bioenergy, carbon capture and storage, smart grid, water supply, usage and
management, waste management, hydro energy, geothermal and biofuel, to name a few.

15 Climate Change Scenarios – Implications for Strategic Asset Allocation


7. Climate policy is a significant contributor to
portfolio risk: Understanding the exposure of
Highlights Figure 5
Contribution to risk for representative portfolio mix
a portfolio to the underlying return drivers is
Equity risk premium
a key component of strategic decision making, 10%
1% Credit risk premium
which is what Figure 6 attempts to measure
through incorporating TIP™ factor risks alongside 5% Illiquidity premium
more traditional risk factors for a representative Technology
portfolio. The existence of risk exposure does Policy
not necessarily imply lower returns, as exposure 12%
can be associated with superior returns under
different market conditions. The aim is to
unravel the sources of portfolio risk and diversify 72%
across the return drivers, as opposed to simply
diversifying between asset classes.
Source: Mercer
Using Mercer’s proprietary Growth Portfolio Toolkit
(GPT), the example is calculated on a hypothetical
but representative portfolio of a typical asset
mix, with allocation of 34% developed large-cap 8. Allocation to sustainable equities, efficiency/
renewable assets, timberland and agriculture
equities, 13% emerging-market equities, 18%
land could improve portfolio resilience: Below is
global government bonds, 26% investment-grade
an illustrative example of the potential impact of
credit and 9% property.7 As can be seen, most of
these asset-class sensitivities on a portfolio mix,
the risk comes through the ERP, as the portfolio
based on optimisation to a nominal return of 7%9
has a high exposure to equities. This can be
that allows for allocation to a wider set of assets.
improved by allocation to a wider range of assets,
As can be seen, in the Delayed Action and Stern
as we will see later in this report.
Action scenarios a sizeable allocation to some of
the climate-sensitive assets (up to 40% of the total
The results show that the climate policy (P)
portfolio) is suggested. Opportunistic investments
factor of the TIP™ framework contributes 10% to
in the Regional Divergence scenario will also be
portfolio risk in this example, with technology
beneficial in the leading regions. Importantly,
(T) contributing just over 1% risk. Impact risk (I)
the risk associated with each scenario varies,
does not appear as a contributor to risk. This can
too, reflecting the higher level of uncertainty
be explained by the small allocation to climate-
associated with the Delayed Action scenario (14%
sensitive assets included in this example that have
risk) compared to the Stern Action scenario (9%
a higher sensitivity to impact risks (real estate,
risk). Climate Breakdown is quite similar to the
infrastructure and commodities), along with the
default case, as it is essentially BAU out to 2030,
evidence pointing to a lower variability in the
although future risks will increase dramatically in
impact risk factor to 2030 (with risks increasing
Climate Breakdown beyond 2050 – hence, a longer
considerably beyond 2050).8
horizon would produce more notable differences.

7 The approach underpinning the growth portfolio toolkit and factor risk approach to asset allocation are explained in the Methodology section (on page 93). Also see
Hawker G. “Diversification: A Look at Risk Factors” (2010), available at http://www.mercer.com/referencecontent.htm?idContent=1378620. For further explanation of
the impact risks, please refer to “Mapping Evidence to Scenarios” on page 75.
8 For further explanation of the impact risks, please refer to “Mapping Evidence to Scenarios” on page 75.
9 The chart shows the optimal portfolio to target nominal return of 7% in each scenario compared to the neutral scenario that does not take climate-risk impacts into
account. Risk refers to the standard deviation in returns. The results should not be used to imply that the most appropriate portfolio to meet these objectives is exactly
as shown. This will depend on factors such as an institution’s existing asset mix, cash rate for the country in which the investor is based, funding position, degree of
risk appetite, investment restrictions and any changes to the assumptions made for risk/return and correlations that may be considered appropriate and potentially
have a significant impact on results. For example, while infrastructure is not included within the allocations shown in the chart, an allocation to infrastructure may be
appropriate based on the rationale provided in this report and the specific opportunities available for investment.

16
Highlights Figure 6
Portfolio to target 7% (nominal) return
100% Cash
90%
Sovereign fixed income
80%
Credit (investment grade)
70%
Developed equity
60%
Emerging market equity
50%
Private equity (including
40%
renewables)
30% Sustainability/renewable
20% themed equities
10% Timberland/agriculture

0% Real estate (core)


Default Regional Delayed Stern Climate
(11% risk) Divergence Action Action Breakdown
(11% risk) (14% risk) (9% risk) (12% risk)

Source: Mercer

9. The EU and China are set to lead the low-


carbon transformation: The regions that are best
costs will need to be absorbed. Grantham LSE/
Vivid Economics point out that the physical
placed to lead the climate change transformation impact costs, as well as the policy adjustment
are those that pre-emptively find alternative costs, will rise substantially in the Climate
sources of energy, improve efficiency, reduce Breakdown scenario beyond 2050 in the absence of
carbon emissions and invest in new technology. any action.
Indicators of current and future investment flows
and policy measures out to 2030 suggest that
the “leaders” are likely to be the EU and China/
East Asia (see Table 4, with sensitivity at the
top and direction denoted by the colour). The
potential for low-carbon transformation in the
US is also significant in the best-case scenario of
Stern Action, but a political impasse on climate
change suggests it may lag in the other mitigation
scenarios, with “improver” countries, including
Japan and India/South Asia, coming through.

While the “do nothing” (Climate Breakdown)


scenario may appear to have lower risk than the
Delayed Action scenario across the regions, that
is because this study focuses on the investment
impacts over the next 20 years when the policy

17 Climate Change Scenarios – Implications for Strategic Asset Allocation


Highlights Table 4
TIP™ factor risk sensitivity and direction of impact for regions
= Positive; = Neutral; and = Negative in terms of the direction of the impact for investments for each region.

TIP sensitivity EU US Japan China/East Russia India/South


Asia Asia

Sensitivity Moderate High Moderate High Moderate Moderate

Regional Divergence

Delayed Action

Stern Action

Climate Breakdown
Source: Mercer assessment as per aggregate estimates, using T, I and P data available at the regional level. Direction of impact derived through a qualitative process.

10. Health impacts and population migration risks


are underestimated: These risks can potentially
change, the sensitivity of asset classes and regions
to these drivers, and the uncertainties that remain,
have an impact on long-term liabilities opening the way to further debate and discussion
and affect assumptions around mortality among investment decision makers.
rates. At present, the evidence available is
not sufficiently strong to draw meaningful Given the high level of uncertainty associated with
conclusions. The health effects will be both climate change, we caution against optimising
positive and negative, and the timing in which portfolio holdings to any one scenario presented in
they will become pronounced is uncertain. The this report. Actions to consider:
research on population migration impacts is
sporadic and qualitative, and further research
will be required to evaluate the potential
1. Understand the risks associated with climate
change and embed these into asset-allocation
impact on pension fund liabilities. Grantham policies. Monitor the evidence related to climate
LSE/Vivid Economics highlight that the existing change in terms of technology, impacts and
studies omit potentially important sources of policy, and discuss what features of the climate
mortality, including malnutrition and deaths scenarios are emerging and what this means for
from extreme events. So they are likely to your investments. This could be built into your
underestimate the increases in illness and annual strategic review and risk management
death between now and 2050. assessments.

Actions for institutional investors to 2. Evolve and transform portfolio mix. Rather than
optimising to any one scenario as presented in
consider
this report, investors could consider a gradual
Institutional investors can respond to the findings of rebalancing of a portfolio towards climate-
this study in a number of ways. The most important sensitive assets that are also tilted towards the
step will be to consider climate change in strategic sustainability theme across infrastructure, private
discussions of long-term investment risks and equity, real estate, timberland and agricultural
opportunities. The framework is not intended to land. This could help to diversify across the
provide a simplistic “tick box” solution for investors to sources of investment risk (including climate
apply in a mechanistic way but to help provide a better change) and improve portfolio resilience across
understanding of the driving forces behind climate the mitigation scenarios.

18
requests for improved disclosure of emissions
3. Allocate to sustainable assets. An additional
response might be an allocation to sustainable levels, environmental impact assessments, as well
investments across both listed and unlisted assets. as full disclosure and reporting of sustainability
This could be viewed as a hedge against some management policies and practices. This can be
of the risks around climate change, particularly undertaken collaboratively through initiatives
climate policy. The risks and opportunities within such as the Carbon Disclosure Project, the
each asset class, as highlighted in this report, Water Disclosure Project, the UN Principles for
could be used as an initial guide for the selection Responsible Investment, or through investor
of the type of investments that might feature in a groups such as the Institutional Investors Group
well-diversified portfolio. on Climate Change (in Europe), the Investor
Network on Climate Risk (in the US) and the
Investor Group on Climate Change (Australia/
4. Consider a wider pool of passive options. Where
portfolios are passively managed, consider New Zealand), to name a few. It could also be
investing in a wider pool of products against undertaken through third-party engagement
different (environmental) indices to better capture agencies, via fund managers that are delegated
the potential upside and/or help mitigate the risks with the management responsibility or, where
of climate change. Passive equity investors should the assets are managed internally, through asset
consider the index constituents and the weighting owners, who can engage directly with investee
attached to sustainability issues when considering companies on these issues.
benchmarks for their investments. They can also
exercise their ownership rights through voting 7. Engage with policymakers. This study showed
that climate policy uncertainty is a notable
and engagement on climate-change issues, either
directly, through third-party agencies or via the source of risk for investors over the coming 20
provider of the passive index product, where years, contributing as much as 10% to risk for
appropriate. Under both the Delayed Action and a representative portfolio. Stretching further
Stern Action scenarios, for example, an allocation into the future, the longer the policy delay, the
to sustainable equities appeared as part of the higher the impact costs will be for investors. It
portfolio mix. is therefore crucial for institutional investors to
engage with policymakers on the specific details
of policy plans and measures as part of their risk
5. Engage with active fund managers. This will help
to ensure that your portfolio is better positioned management process, to help protect and enhance
for responding to the uncertainties in a way that the long-term value of the assets they oversee.
helps reduce the risk of being too late, reactive This should go beyond high-level motherhood
and costly. Ask your fund managers to specify key statements and should be appropriately resourced
criteria and pressure points that they will measure and focused on targeting specific policy measures
and integrate into their investment processes. This at the local and global levels, to actively manage
might include an ongoing assessment of climate the policy risk that climate change produces.
policy developments, cost-of-carbon scenario
analysis, the impact of technology flows on risks 8. Support ongoing research. Consider areas
for further research and look for collaborative
and opportunities, and an evaluation of any
possible risks from climate damage, including on opportunities to support these endeavors with
assumptions regarding expected returns such as academics, policymakers and relevant experts.
the ERP. Some ideas include the following:

Continue to evaluate the impact of climate


6. Engage with companies. Institutional investors n

should engage with companies in which they are change on strategic decision making. This
invested on climate risk management issues to study developed a framework with which to
proactively manage the risks. This will include examine climate change and its potential

19 Climate Change Scenarios – Implications for Strategic Asset Allocation


impact on long-term risks/returns across asset
classes and regions. However, institutional
investors need to apply the results to their
portfolios to evaluate the risks they face and
internalise the framework into their decision-
making processes. This will also involve
supporting the development of new tools and
approaches as the climate change data and
evidence changes over time.

n Spend time exploring the best way to build


exposure. The implementation of the findings
of this study at the asset-class and regional
levels needs to be carefully considered in terms
of the right vehicle to use and the preferable
approach to take. It is essential for institutional
investors to spend time considering ways
to allocate to the opportunities across the
asset classes in a cost-effective and prudent
manner. This means exploring the costs and
benefits of investing in funds, fund of funds,
co-investments or public-private sector
partnerships, and/or making direct investments
in projects.

n Monitor the scientific evidence on the


physical impacts of climate change. The
range of uncertainty in projecting long-term
climate impacts is wide ranging due to many
unknowns in the causal chain of climate
impacts. For example, if tensions over water
resources increase due to droughts, the result
could be social pressures leading to changes
in governments, migration and conflict. Costs
could easily be much greater than the range
estimated in this report. Investors therefore
need to monitor new scientific evidence and
social pressures related to climate change.

n Research the impact on pensions of


population migration. This study highlighted
the lack of research on the potential impact
of climate change on population migration,
including what regions will be most affected,
how governments are likely to respond and
what implications may arise for pension funds
around the world. Research of this kind, with
the participation of the actuarial community,
would enable better analysis of the impact of
climate change on liabilities than is currently
available.

20
Messages for policymakers

The key messages for policymakers from this study are:

1. Policy is crucial for mobilising capital. The policy


environment is one of the key factors that investors
will consider when deliberating about climate
change, as it will be an important signal for future
investment in technology-related opportunities
and also potential risks associated with changes
to the physical environment. Indeed, the risk that
investors will attach to such investments under
a clear and well-coordinated policy framework is
considerably lower than a late or disparate policy
approach.

2. Make policies clear, credible and coordinated.


Policy design needs to be clear, credible and well-
coordinated internationally to attract institutional
assets and to help reduce risk premiums assigned
to riskier investments. A high level of policy
uncertainty will increase volatility and lead
investors to demand a higher risk premium on their
investments than would otherwise be the case.

3. Delay now, pay (more) later. Our Delayed Action


scenario predicts that most core assets will suffer
as a result of unforeseen and dramatic policy
action. If this situation emerges, investors will
demand a higher cost of capital in the future as
risk aversion rises. The investment impact of this
scenario is negative for all countries/regions – as
the future cost of carbon rises, the longer the delay
will be, meaning there will be no long-term winners
from a delayed response (although some countries
may pose a greater investment risk than others).
Many investors may be reluctant to invest in low-
carbon opportunities until the policy framework
is in place, potentially increasing the required rate
of return on such investments in the intervening
period.

21 Climate Change Scenarios – Implications for Strategic Asset Allocation


22
Overview of
investment impacts
This section presents the highlights of the
investment impacts, with further details
provided on each asset class and region in
Sections 2 and 3, respectively. This will be
presented in seven parts:

1. Source of investment risk across the


climate scenarios
2. Sensitivity of assets to investment risk
3. Impact on the equity risk premium (ERP)
4. Estimates of TIP™ risk factors
5. Sensitivity of assets to the TIP™ risk
factors
6. Sensitivity of regions to the TIP™ risk
factors
7. Quantitative analysis

23 Climate Change Scenarios – Implications for Strategic Asset Allocation


Source of investment risk across the climate
scenarios
Table 1 presents Mercer’s interpretation of how  ction scenario, in which inflation and interest
A
the different sources of investment risk vary for rates increase due to an (unanticipated) carbon
each climate scenario. This was derived through price shock.
a qualitative process that followed a series of
discussions between the Mercer project team and n  arket risk factors – the ERP and volatility are
M
in-house asset-class experts, the Research Group expected to change in the more extreme mitigation
and the rest of the project group participants (see scenarios, in which policy changes and the degree
Methodology, on page 93, for further details). of transformation in technology increase. This
expectation is based on comparable periods in
The greatest sources of investment risk across the history that have produced a significant difference
climate scenarios are expected to come through between historical risk premiums over time due
changes to climate change risks and (to a lesser to transformative events (see discussion below for
extent) market risks, rather than through fundamental further details).
risks. A few additional observations:
n  limate change risk factors – changes in technology
C
n  undamental risk factors – the fundamental
F and climate policy are the driving forces behind
risks are not expected to change for most of the the differences across the climate scenarios.
climate scenarios. This is largely based on the Where policy measures are anticipated by the
Grantham LSE/Vivid Economics finding that the market (Stern Action), the result is generally more
macroeconomic impacts become more pronounced favourable for the assets most sensitive to climate
beyond the time horizon of this study – that is, change. A more detailed discussion of the sources of
beyond 2050. The exception to this is the Delayed risk around climate change is provided below.

Table 1
Impact of scenarios on sources of investment risk

Scenario Fundamental Market


Climate change factors
factors factors

Economic cycle ERP Technology Impact Policy


Inflation Volatility

Regional Unchanged Higher High dispersion of Higher risk of future Higher uncertainty
Divergence volatility capital inflow into low- impact costs due to and potentially higher
carbon investments; slower reduction in reward for some
leading countries emissions assets due to regional
include the EU and disparity in climate
China policy

Delayed  igher
H Higher Business as usual Higher risk of future Higher uncertainty
Action inflation volatility (BAU) investment in impact costs due to around policy until
Higher interest Lower low carbon until 2020 delay in policy response 2020, then dramatic
rates realised ERP when policy measures U-turn reduces policy
stimulate flows uncertainty

Stern Unchanged Lower Clarity on climate policy Lower risk of future Policy clarity at the
Action volatility stimulates strong capital impact costs due to global level reduces
Higher flows into low-carbon reduction in emissions investment uncertainty
realised ERP solutions

Climate Unchanged Unchanged; Higher risk attached to Higher impact risks due BAU climate policy
Breakdown risk of low-carbon technology to lack of policy action, (unchanged from
higher investments due to rising future costs and today’s measures)
volatility policy inaction market pricing in future
policy shift
Source: Mercer

24
Sensitivity of assets to investment risks

The sensitivity of each asset class to the different n In contrast, real estate, infrastructure, private
sources of investment risk is presented in Table 2, equity, sustainable equities, efficiency/renewables
where the asset classes are located according to and commodities are highly sensitive to climate
whether they have a high or very high sensitivity change factors.
to each source of risk. Some of these risks can be
quantified, such as the ERP and volatility. However, To put it simply, this means that portfolios that are
some risks cannot be quantified but are still important dominated by listed equities and bonds may not be as
to consider as part of the risks associated with an sensitive to climate change, which may be a positive
investment strategy.10 A few highlights from Table 2: outcome under a “no mitigation” scenario such as
Climate Breakdown (which is also the least likely
n Listed equities, government bonds and investment scenario). For all the other scenarios where some
grade credit all have high sensitivity to fundamental degree of mitigation will occur, there will be portfolios
risk factors but not to climate change factors. with a low allocation to assets that are sensitive to
climate change may be less resilient in terms of both
the risks and the opportunities.

Table 2
Assets with high or very high sensitivity to investment risks

Fundamental factors Market factors Climate change factors

Economic cycle ERP Technology


Inflation Volatility Impact
Policy

Listed equities Listed equities Real estate


Emerging equities Private equity Infrastructure
Government bonds Infrastructure Private equity
Emerging debt Real estate Sustainable equities
Investment-grade credit Efficiency/renewables
Commodities Commodities
Source: Mercer

10 See “Diversification: A Look at Factor Risks”, available at http://www.mercer.com.br/referencecontent.htm?idContent=1399985.

25 Climate Change Scenarios – Implications for Strategic Asset Allocation


Impact on equity risk premium

Broadly defined, the ERP represents the compensation increases the prospect of potentially large and
for taking on equity risk versus a risk-free rate. The persistent event risks related to water or resource
notion of the ERP is widely used in finance models scarcity (creating geopolitical tensions) and large-
and also features as an input into the way Mercer scale weather-related events. While such climate-
develops some of its asset-class assumptions. Hence, related events cannot be predicted, the scientific
it is important to consider whether the climate- evidence suggests that the incidence and magnitude
change scenarios might impact the ERP and, if so, of such events are likely to increase, owing to
in what way and by how much. The following ERP climate change, in which case the ERP will need to
discussion focuses on realised returns for an existing reflect that risk.
portfolio of assets at a future point in time, which is
most commonly referred to as the Historical Equity Damodaran considers three approaches to estimating
Premium. This is because the study is evaluating the risk premiums, including the survey approach, the
outcome/consequence of different climate scenarios historical premium approach and the implied premium
for an existing portfolio of assets, starting from today approach. For climate change it is not possible to infer
and looking at the outcomes at a future end date (in the premium on the basis of historical or implied
this case 2030). market data, as the market is not yet pricing in
climate risk in a meaningful way, and additionally,
As Damodaran (2008) summarises,11 the ERP it would not allow us to capture how the ERP might
assumption reflects a fundamental judgement about change under alternative climate scenarios. Hence,
how much risk we see in the market and what price we have formulated a possible directional impact
we attach to that risk. Some of the key determinants of on ERP assumptions based on a qualitative in-house
the ERP and a possible link to climate change factors assessment within Mercer and consultation with the
are: project group on how the ERP might change under the
climate scenarios.12
n Overall risk aversion – This may increase in
scenarios where climate change increases the In addition to this analysis we have also drawn
overall level of uncertainty, as well as in situations from lessons in the past – in particular, the findings
where there is a period of transformation in the of Dimson et al (2003)13 on how the historical ERP
economy that is costly and unanticipated. changed from the first half of the 20th century
compared to the second half. In brief, the authors
n The degree of uncertainty – The sources of found that the ERP versus Treasury bills was 4.1%
uncertainty associated with each climate scenario in the first half of the 20th century and 7.7% in
relate to the degree of technology development the second half. In other words, there was a 3.6%
and deployment, climate policy transparency and difference in the ERP between the first and second
coordination, and physical impact risks. halves of the century. This was based on 16 countries
and 102 years of data.
n The level and reliability of available information –
Poor transparency on climate-change-related risks There are obviously a range of possible explanations
combined with increased uncertainty about how for this divergence, but the authors deduced that
to interpret new information could make investors periods of turmoil and economic/political uncertainty
less certain about the future and lead to higher risk – such as the first half of the century, with two world
premiums. wars and the Great Depression – were associated with
lower realised return on riskier assets. In contrast,
n Catastrophic/event risk – Past examples of periods of progress and technological advancement –
catastrophic risks that can cause dramatic drops in such as in the second half of the century, with the IT
wealth include the Great Depression of 1929–1930 in revolution and increased levels of productivity – are
the US and the collapse of Japanese equities in the associated with higher realised return on riskier assets.
1980s. It is not inconceivable that climate change

11 Damodaran A. (2008). Equity Risk Premiums (ERP): Determinants, Estimation and Implications – The 2010 Edition, available at http://papers.ssrn.com/sol3/papers.
cfm?abstract_id=1556382, accessed 11 January 2011.
12 The results represent the broad consensus that emerged across the project partner members.
13 Dimson E, Marsh P and Staunton S. “Global Evidence on the Equity Risk Premium”, Journal of Applied Corporate Finance, Vol. 15, No. 4, Summer, 2003, pp. 8–19.

26
This divergence is important for interpreting the Pulling all of these inputs together, during periods
impact of the climate-change scenarios on the possible of higher uncertainty around climate change, lower
ERP outcomes. realised returns on riskier assets are more likely to
emerge than under more optimistic scenarios of
n The period of uncertainty and turmoil of the first positive economic transformation. The macroeconomic
half of the century has similarities to the Delayed impacts have also informed these conclusions, where
Action scenario, which may also bring a period higher interest rates and inflation associated with
of destruction that is not fully anticipated by the the Delayed Action scenario also support a potential
market and, hence, costly for some investments. We decline in the ERP in that scenario. The final column
therefore expect that this will lead to lower realised in Table 3 is constructed on that basis and refers to
returns in 2030. the realised return in 2030 on an existing portfolio of
assets.
n In contrast, the second half of the century is more
akin to the Stern Action scenario, where a period of
transformation involves significant new advances in
technology, with supportive and transparent policy
creating efficiency gains and a positive environment
for some investments, with lower volatility expected.

Table 3
Determinants of the ERP across the climate scenarios

Scenario Degree of risk Degree of investment Reliability Event risk Change of realised
aversion uncertainty of available ERP in 2030
information

Regional Unchanged overall Varied by regions with Deteriorates in Increases Higher volatility
Divergence but varies by leaders and laggards laggard regions,
region creating higher leading regions
uncertainty overall improves

Delayed Increases Higher uncertainty before Deteriorates then Increases Lower realised ERP
Action policy changes which improves post (driven by lower
are not anticipated; policy shift returns and higher
uncertainty declines volatility)
following policy measures

Stern Unchanged Lower uncertainty Improved as Lower Lower volatility


Action around climate policy businesses are
due to transparency required to
that is coordinated and report audited
anticipated emissions data

Climate Broadly unchanged Low uncertainty Unchanged Increases but Unchanged with the
Breakdown to 2030, increasing until 2050, but then from 2050 risk of dramatic shift
by 2050 increasing, possibly onwards further into future
abruptly intensifies
Source: Mercer

27 Climate Change Scenarios – Implications for Strategic Asset Allocation


Estimates of the TIP™
factor risks
To better analyse the investment impact of climate
change, Mercer developed the TIP™ risk factor
framework to examine which factors drive asset-class
returns across the following three areas:

n Technology (T) – broadly defined as the rate of


progress and investment flows into technology
related to low carbon and efficiency, which are
expected to provide investment gains

n Impacts (I) – the extent to which changes to the


physical environment will affect investments
(negatively)

n Policy (P) – the cost of policy in terms of the change


in the cost of carbon and emissions levels that result
from policy, depending on the extent to which it is
coordinated, transparent and timely

The results of the TIP™ factor risk estimation are


summarised in Table 4 (on page 29), expressed both
as the future value of each factor in 2030 as well as
the annual change to 2030 for each of the T, I and P
factors in $US trillion. The total value for each region
is also presented in the table. The regional values
do not compute to 100% of the global value, as the
analysis represents only the major countries that have
comparable data.

For impact costs under the Climate Breakdown scenario,


many of the developing economies are the worst hit,
which explains the gap in the computations at the
regional level for that factor, as they are not all included
in Table 4. Nevertheless, the regions included represent
the largest in terms of low-carbon investment flows and
carbon emissions levels.

As the results illustrate, we expect that there may be


significant variability in the TIP™ factors across the
scenarios, with a particularly wide dispersion coming
through the policy and technology factors.

28
Table 4
Value of the TIP™ factors across the climate scenarios

Scenario Cumulative Cumulative Change in Change on per annum basis to 2030


value of value of value of Policy (US$)
Technology Impacts to to 2030 (US$
Technology Impacts Policy
to 2030 (US$ 2030 (US$ trillion)
$US bn $US bn $US bn
trillion) trillion) % total by
% total by % total by region
region region

Regional Divergence $3.9 $3.0 $4.8 $194 $149 $240


– EU 25% 9% 8%
– US 17% 32% 23%
– Japan 3% 4% 5%
– China/E Asia 35% 21% 25%
– Russia 1% 7% 24%
– India/S Asia 6% 19% 15%

Delayed Action $3.7 $2.3 $8.1 $183 $112 $405


– EU 19% 9% 14%
– US 27% 32% 19%
– Japan 5% 4% 4%
– China/E Asia 29% 20% 42%
– Russia 4% 7% 8%
– India/S Asia 10% 19% 13%

Stern Action $5.2 $1.5 $2.6 $259 $76 $130


– EU 19% 8% 11%
– US 26% 32% 23%
– Japan 4% 4% 5%
– China/E Asia 26% 20% 35%
– Russia 4% 6% 10%
– India/S Asia 9% 19% 16%

Climate Breakdown N/A as the $3.7 $240 bn Baseline $186 $12


– EU technology 4% 13%
factor is
– US 14% 21%
calculated as
– Japan incremental 2% 4%
– China/E Asia investment flows 8% 43%
– Russia in addition to 3% 7%
investments that
– India/S Asia 10% 12%
would take place
under BAU
Source: Mercer calculations based on International Energy Agency (IEA) World Energy Outlook 2009 data and Grantham LSE/Vivid Economics reports
Net present value discounted at a rate of 3%. The country and regions included in this analysis are a partial representation of the global economy – hence the reason
the totals by region do not compute to 100% for each TIP™ factor. However, in all scenarios the countries included in this study represent around 90% of the total
TIP™ risks, as they include the largest markets in terms of investment and emissions levels. Other countries were omitted due to a lack of comparable data to estimate
the TIP™ values, but future updates will aim to incorporate markets such as Latin America (Latam), South Korea, Africa, and the Middle East and North Africa (MENA)
as data become available.

29 Climate Change Scenarios – Implications for Strategic Asset Allocation


Technology – size of future investment flows n This equates to an increase of $180 bn to $260 bn
pa from current levels, with the highest end of the
n The dollar value of this factor can be interpreted spectrum representing the Stern Action scenario.
as a measure of the future private-sector low- This would be additional investment on top of
carbon investment flows under different climate recent flows, which have fluctuated between $40 bn
scenarios, where a higher technology value indicates and $170 bn pa over the past five years.14
a higher level of investment. It is important for
investors to have a sense of the low-carbon n On a regional basis, the biggest differences between
investment flows across the climate scenarios as the countries are for the Regional Divergence and
an indicator of the potential depth of the pool of Delayed Action scenarios. As Figure 1 highlights,
investment opportunities. the “leaders” in technology under the Regional
Divergence scenario are expected to be China/East
n The variability in this factor across the scenarios is Asia and the EU. The US, “mature but contracting”
high, falling in the range of 13%–60% pa higher than in this scenario, nevertheless represents a
a BAU level of investment. substantial market in terms of depth. Under
Delayed Action, the investment flows are expected
n Additional cumulative investment in efficiency to be slightly lower overall due to the delay in
improvements, renewable energy, biofuels, and getting started, with the US playing catch-up along
nuclear and carbon capture and storage could with some of the “improving” regions, such as Japan
expand in the range of $3 tr to more than $5 tr by and India/South Asia.
2030 across the mitigation scenarios examined in
this study.

Figure 1
Cumulative Investment in Technology to 2030 by region
Regional Divergence

Delayed Action

1600

1400

1200

1000
$US bn

800

600

400

200

0
EU US Japan China/East Asia Russia India/South Asia
Source: Mercer computations based on IEA WEO (2009), as defined in the Methodology (see page 93)

14 UNEP, SEFI and Bloomberg New Energy Finance (2010).

30
Impacts – cost of physical climate change the earlier observation in this report that the current
impacts response from policymakers puts us closer to the
Regional Divergence or Delayed Action scenarios. The
n Some investments may be directly affected by greater the policy commitment and the earlier and
rising risks of climate-change-related events with more coordinated the action (as for Stern Action),
an impact on the value of their assets. As such, the lower the resulting impact costs will be. Under
the economic cost of the physical changes is an a Delayed Action scenario, despite the high level of
important variable for investors to monitor over time effort and expense associated with climate policy,
as part of their overall risk management process. the impact costs are almost as high as those for the
Climate Breakdown scenario. This is because the
n The cumulative economic cost of changes to the policy action is late and also unanticipated (hence,
physical environment, health and food security is more costly).
estimated to be in the range of $1.5 tr to $3.7 tr to
2030 across the climate scenarios – and as costs rise, n On a regional basis, the impact costs increase under
the greater the delay and the less well-coordinated scenarios where progress on policy and technology is
the policy response will be. This represents $70 bn to slow, late or inefficient. Figure 2 plots the percentage
$180 bn pa to 2030. of GDP for the two most extreme scenarios in terms
of climate impacts – Stern Action and Climate
n The current international policy discussion to Breakdown. The impact costs rise considerably
commit $100 bn pa, with priority on adaptation and in India/South Asia under a BAU scenario, where
residual damage costs, falls short of covering the physical impact risks in the time horizon of this
impact costs under all the scenarios except for the study are the highest of the countries analysed.
most optimistic one – Stern Action. This confirms

Figure 2
Adaptation and residual damage costs by region
Stern Action

Climate Breakdown
0.9

0.8

0.7

0.6
% GDP in 2030

0.5

0.4

0.3

0.2

0.1

0
EU US Japan China/East Asia Russia India/South Asia

Source: Grantham LSE/Vivid Economics and Mercer Impact factor calculations, as defined in Methodology (see page 93)

31 Climate Change Scenarios – Implications for Strategic Asset Allocation


Policy – change in cost of emissions n Based on the projected trajectory of emissions
across the scenarios, the change in the (implicit
n The degree to which climate-related policy action or explicit) cost of carbon could increase by $2
takes place and is anticipated by investors will trillion to $8 trillion from 2010 to 2030 across the
be the key factor to consider when evaluating the mitigation scenarios depending on the policy
investment impacts of climate policy. In a situation approach taken, where the upper end represents
where credible policy and a higher carbon cost are the Delayed Action scenario. Policy delay therefore
fully anticipated, the impact may be positive for the represents a substantial additional carbon cost for
highly sensitive assets, as emissions and uncertainty the market to absorb.
around policy both will be reduced (and vice versa).
n On a regional basis, the greatest differences emerge
n The cost of climate policy is a function of how in the Delayed Action and Stern Action scenarios,
quickly the policy action is taken, the level of where policy delay hits China particularly hard given
emissions associated with each scenario and its rising emission levels and the resultant higher
the cost of carbon. The cost of carbon emissions cost of carbon. The higher cost in Delayed Action
increases the longer the policy is delayed, and the versus Stern Action suggests that policy risk is a key
less well-anticipated and coordinated the policy factor for investors to take into account across all
action is. Grantham LSE/Vivid Economics estimated regions.
that the cost of carbon for mitigation scenarios could
be $110/tC02e–$220/tC02e by 2030, compared with the
current EU ETS price equivalent of around $15/tC02e.
These costs may be explicit in the market or implicit
due to policy measures that affect operating costs
outside of emission trading schemes.

Figure 3
Policy cost comparison
Delayed Action

Stern Action
1200
$US bn cost of emissions in 2030

1000

800

600

400

200

0
EU US Japan China/East Asia Russia India/South Asia
Source: Mercer Policy factor calculations

32
Sensitivity of assets to the TIP™ risk factors

The “sensitivity of assets” refers to the degree to n Equities and fixed income have a lower sensitivity to
which the underlying risk/return drivers of assets climate change across the scenarios. This is because
are sensitive to the climate-change TIP™ factors. The the differences are expected to be greater at the
process of deriving these TIP™ factor sensitivities sector and regional levels rather than at the asset-
was largely a qualitative exercise, developed jointly class level. It is also due to a combination of being
between climate specialists and asset-class specialists more sensitive to other sources of risk (fundamental
both within Mercer and across the project participants. factors) and the shorter time horizon in terms of
Table 5 summarises the results, the key conclusions of asset pricing compared with some of the more
which are highlighted as follows: climate-sensitive assets.

n The assets whose underlying risk/return drivers n The exception to this would be investments in
are most sensitive to climate change include sustainability-themed listed equities and efficiency/
infrastructure, private equity, real estate, renewables, where the sensitivity to climate
sustainability-themed listed equity, efficiency/ change is obviously much higher and pre-emptive.
renewables, timberland, agricultural land and carbon. Consequently, these assets could help investors
As a consequence, these assets will likely capture capture more of the upside and protect against
the greatest opportunities under the mitigation unforeseen risks than a traditional equity and bond
scenarios and also pose the greatest risk under the portfolio would do. They would also expose investors
Climate Breakdown scenario. to the downside risk of a “no-mitigation” scenario
such as Climate Breakdown, although that is also the
least likely scenario.

Table 5
Sensitivity of asset classes to climate change risks
Sensitivity of the impact: where L = Low; M = Moderate; H = High; VH = Very high sensitivity to the combined climate change factors.

Direction of the impact: where = Positive; = Neutral; and = Negative. Agriculture = agricultural land; RE = real es
estate;
Infra = infrastructure; EME = emerging-market equity; EMD = emerging-market debt; LBO = leveraged buyout; VC = venture capital.

Listed equities Fixed income Commodities RE Private equity Infra


Efficiency/renewables

Efficiency/renewables

Efficiency/renewables
Sustainable equity

Agricultural land
Inv grade credit

Core, unlisted
Global equity

Global fixed

Timberland

Unlisted
EMD
EME

LBO

VC

Sensitivity L M H VH L M L H H H M H VH H VH

Regional Divergence

Delayed Action

Stern Action

Climate Breakdown
Source: Mercer. Sustainable equity = broad multi-themed listed equity companies that generate a substantial proportion (typically more than 25%) of their earnings
through sustainable activities. Efficiency/renewables assets = both listed/unlisted sustainability-themed assets whose core activities are theme specific and more
concentrated in terms of exposure than are broad sustainability equity. This includes (but is not limited to) energy efficiency, low energy transport, renewable energy,
bioenergy, carbon capture and storage, smart grid, water supply, usage; and management, waste management, hydro energy, and geothermal, to name a few.

33 Climate Change Scenarios – Implications for Strategic Asset Allocation


Sensitivity of regions to the TIP™ risk factors

Asset-allocation decisions are typically conducted n New market opportunities in the development and
at the top-down level across regions, at most trade of technologies and low-carbon solutions
differentiating between local, emerging and opening up new markets and industries for regions
developed markets. However, the potential risks that develop capabilities in this area
and opportunities at a more disaggregated level may
be informative for some asset classes where regional n Future cost of carbon exposure being expected to
disaggregation can be implemented (for example, rise over time internationally – the uncertainty
equities, sovereign fixed income, private equity, relates more to the pace and level of change;
infrastructure and real estate). hence, it may become more costly for high-emitting
countries to redress their carbon liabilities over time
For long-term institutional investors, the wider
socioeconomic trends that are driving the shift away The regions examined in this report are limited
from high-carbon activities are important factors to those for which comparable data sources were
to take into account in considering the sensitivities available within the TIP™ factor risk framework.
of regions to the TIP™ risk factors, some of which Table 6 summarises the outcome in terms of the
include: relative importance of the climate factor risks for each
region.16
n Increased energy security in a world of depleting
resources and growing population that is spurring
sharp increases in energy demand15

n Increased energy efficiency across industries, leading


to improved international competitiveness in the
long term

Table 6
Sensitivity of regions to climate change risks
= Positive; = Neutral; and = Negative in terms of the direction of the impact for investments for each region.

TIP™ sensitivity EU US Japan China/East Russia India/South


Asia Asia

Sensitivity Moderate High Moderate High Moderate Moderate

Regional Divergence

Delayed Action

Stern Action

Climate Breakdown
Source: Mercer assessment as per aggregate estimates, using T, I and P data available at the regional level. Direction of impact derived through a qualitative process.

15 Lloyds 360o Risk Insight White Paper, “Sustainable Energy Security, Strategic Risks and Opportunities for Business” (Chatham House, 2010).
16 The calculation and sources of data for all estimates at the regional level are further explained in the Country and Regional Impacts section (page 69) and the
Methodology section (page 93) of this report.

34
The analysis concludes that countries that act pre-
emptively in finding alternative sources of energy,
improving efficiency, reducing carbon emissions and
investing in new technology may benefit from the
transformation that will take place in the form of
higher investments with less uncertainty than laggard
countries and hence, potentially will be more attractive
for long-term investors. Countries that avoid or delay
the transformation process may face a more uncertain
future with regard to higher carbon mitigation costs
and therefore potentially pose a higher risk for long-
term investors. Moreover, a delay in policy action may
cost all regions, as it pushes up the adjustment costs
that will be required to reduce emissions in the future
– hence, there are no winners from delay. Overall, this
increases the risks for global investors.

While the “do nothing” (Climate Breakdown) scenario


appears to be less negative across the regions than
the Delayed Action scenario, as explained previously,
that is due to the horizon of this study focusing on
the impacts over the next 20 years. Grantham LSE/
Vivid Economics point out that impact costs may
push up the cost of Climate Breakdown considerably
beyond 2050.

It is important to interpret these results with some


caution, as there is substantial variation in the
type of policy measures that different countries
have introduced that may be masked by the overall
assessment presented here. The analysis is also
based on a limited number of countries due to
constraints with comparable data, although those
included are the largest nations in terms of low-
carbon technology flows and carbon emission levels.
Nevertheless, investors need to assess the climate risks
of investments by region on a case-by-case basis and
continually monitor the policy developments at the
regional level.

35 Climate Change Scenarios – Implications for Strategic Asset Allocation


Quantitative analysis results

We have made some additional quantitative Figure 4


assumptions relating to asset-class exposure to each Contribution to risk for representative portfolio mix in ‘default’
of the TIP™ factors based on our qualitative analysis. case
In doing so, we were cognisant that there will
Equity risk premium
always be some subjectivity and uncertainty when 10%
1% Credit risk premium
determining appropriate assumptions. Appropriate
5% Illiquidity premium
assumptions can also vary significantly depending
Technology
on the investor and the purpose for which the
Policy
assumptions are to be used.
12%

The aim of our quantitative analysis is not to suppose


that it can calculate the most optimal portfolio for the
next 20 years based on the climate-change analysis. 72%
We acknowledge that mean-variance optimisation
analysis can be extremely sensitive to assumptions
and is not sufficiently comprehensive to assess all Source: Mercer
characteristics of different investments. Instead, the
analysis is intended to act as a sense check to our In this example, the policy (P) factor of the TIP™
qualitative assessment. framework contributes 10% to portfolio risk, while
technology (T) contributes 1.4% risk. The T factor is
We have used Mercer’s GPT17 risk factor framework comparatively low, as the assets with higher sensitivity
to develop and test the internal consistency of the to technology factors, such as private equity and
assumptions across the asset classes. sustainability-themed assets, are not included in
the example. While the technology risk exposure is
Source of risks expected to be beneficial for portfolio returns, it cannot
be obtained without taking on policy risk, which is a
Understanding the exposure of a portfolio to the necessary cost attached to the technology exposure.
underlying return drivers is a key component of Furthermore, policy risk has some correlation to
strategic decision making, which is what Figure 4 overall market risk, as it introduces additional market
attempts to measure through incorporating TIP™ uncertainty and, hence, is a potential source of
factor risks alongside traditional risk factors. The volatility. Policy risk therefore is embedded within
existence of risk exposure does not necessarily imply all asset classes and not only those that are highly
lower returns, as exposure can be associated with sensitive to climate change. This explains why the
superior returns under different market conditions. policy risk exposure is relatively high in the example
The aim is to unravel the source of portfolio risks and provided.
diversify across the return drivers as opposed to simply
diversifying between asset classes. Impact (I) risk does not appear as a contributor to
risk. This can be explained by the small allocation
The example in Figure 4 is calculated on a hypothetical to tangible assets (infrastructure, commodities, real
but representative portfolio of a typical asset mix, estate) included in this example that have a higher
with allocation of 34% developed large-cap equities, sensitivity to impact risks, along with the evidence
13% emerging-market equities, 18% global government pointing to a lower variability in the impact risk factor
bonds, 26% investment-grade credit and 9% property. compared to the low-carbon technology transformation
and change in carbon costs associated with climate
As can be seen, most of the risk comes through the policy out to 2030. Investors should, however, be aware
ERP, as the portfolio has a high exposure to equities. that the evidence suggests that further into the future
(beyond 2050), the source of portfolio risk attributable
to the impacts factor will increase considerably.18

17 Further information on Mercer’s approach to factor risk and asset allocation is explained in the Methodology section (page 93) of this report. Also see Hawker G.
“Diversification: A Look at Risk Factors”, available at http://www.mercer.com/referencecontent.htm?idContent=1378620.
18 For further explanation of the impact risks, refer to the Mapping Evidence to the Scenarios section (page 75) of this report.

36
Illustration of potential portfolio impact given level of return, or vice versa) under the “default”
or default assumptions.
The following analysis is based on a traditional and
widely used mean-variance analysis approach. While The changes in performance of the assets are
such modelling has a number of drawbacks and, calculated versus this “default” assumption starting
hence, critics (for example, it can be highly sensitive point, which assumes that the climate change factors
to changes in assumptions – it assumes a normal contribute to overall portfolio risk without the benefit
distribution range of returns and that asset-class of providing additional return.
relationships are constant over time and, hence, like
any model, it cannot capture all characteristics of We have summarised part of our analysis in Figure
investments), it can be used to provide some additional 5 (on page 38). The chart shows the portfolios, as
insight into appropriate portfolio structure. A few calculated by the model, that target a nominal return
specific approaches have been used for this analysis: of 7% in each scenario, with lowest standard deviation.
The results should not be used to imply that the most
n The assumptions have been modified to reflect the appropriate portfolio to meet this objective in each
TIP™ factor risks and the sensitivity of each asset scenario is exactly as shown. For example, other sets of
class to these sources of risk, in addition to the more assumptions are reasonable and the most appropriate
traditional drivers of asset-class risk/returns. assumptions are likely to change over time. In
addition, the characteristics of sustainability-themed
n The GPT factor risk framework has been used to test investments will often be unique to each product and
and examine the underlying assumptions to ensure manager and, hence, cannot be represented by generic
that the modifications are internally consistent and asset assumptions in a simplistic way. A different
pass the “common sense” test. set of assumptions may need to be used for assets
depending on the degree to which they are single
n Different climate scenarios have been examined or multi-themed, multi-country, or in the listed or
to provide a sense of how portfolios respond to unlisted space.
different conditions, highlighting the areas of
potential weakness as well as opportunity. Despite these caveats, we believe that the analysis
helps to demonstrate that an increase in investor focus
n Additional types of “sustainability” investments on climate-sensitive assets is likely to be rewarded by
within the asset classes have been included in the a reduction in risk or improvement in return in some
modelling analysis to examine the extent to which of the climate-change scenarios analysed. The precise
an allocation to such investments might improve the areas of focus in each scenario will depend on more
resilience of a portfolio mix to climate change. robust supplementary analysis specific to an investor’s
risk/return profile, product characteristics, timing and
Despite these modifications, we recognise the the assets in investors’ current portfolios.
drawbacks of mean-variance analysis and have
exercised some caution in structuring our assumptions In addition to showing the composition of the
and conclusions on how sustainability themed “optimal” portfolio under each scenario, the chart
asset classes may benefit investors. Our “default” also includes the level of risk for each scenario. The
assumption for this modelling is that climate change risk measure shown is the standard deviation of
does not affect the expected return of investments the returns. The key differences in the amount of
but it does add some additional uncertainty. Because investment risk that is necessary to meet the return
of this, it assumes that sustainable equities and target arise in the Delayed Action and Stern Action
efficiency/renewables (unlisted) may provide the scenarios. Under Delayed Action, more risk is
same return but with higher risk compared with their required because asset returns are expected to be
broader market equivalent assets. It is for this reason lower. In contrast, under Stern Action, less risk is
that these asset classes do not feature in the calculated needed because higher returns are expected from
optimal portfolios (a portfolio that minimises risk for a each asset class.

37 Climate Change Scenarios – Implications for Strategic Asset Allocation


Figure 5
Example of portfolio mix across the scenarios – portfolio to target 7% return

100% Cash
90%
Sovereign fixed income
80%
Credit (investment grade)
70%
Developed equity
60%
Emerging market equity
50%
Private equity (including
40%
renewables)
30% Sustainability/renewable
20% themed equities
10% Timberland/agriculture

0% Real estate (core)


Default Regional Delayed Stern Climate
(11% risk) Divergence Action Action Breakdown
(11% risk) (14% risk) (9% risk) (12% risk)

Source: Mercer

Looking at the analysis in more detail, the greatest n Delayed Action is negative for most asset classes –
portfolio impacts occur for the Delayed Action and albeit, we would expect some improved performance
Stern Action scenarios, with little change under from climate-sensitive assets over time to alter the
Climate Breakdown due to the constraint on the optimal portfolio. This is illustrated by a projected
investment horizon to 2030. For much longer time increase in efficiency/renewables unlisted assets
horizons the investment risks associated with Climate in Figure 5 (which could be obtained via private
Breakdown would increase substantially, driven by equity or infrastructure), timberland and agricultural
higher “impact” factor risks. We believe that the land. Figure 6 (on page 39) translates these new
analysis is explained by the following factors: allocations into a risk factor framework and, as
can be seen, the new asset mix would improve the
n Regional Divergence implies that the backdrop diversification to different risk factors, with a much
for climate-sensitive investments will vary by lower exposure to the ERP and a higher exposure to
region. Hence, while we expect the returns of these illiquidity, technology and policy. Such a portfolio
investments to increase on average, we also expect is better diversified and therefore likely to be more
the uncertainty and variation in returns to increase. robust should climate policy be delayed.
It is for this reason that the quantitative analysis
does not show assets such as timberland, agriculture
land, efficiency/renewables, real estate, sustainable
equity or infrastructure as part of the “optimal”
results in this scenario – because of the higher
uncertainty and sensitivity of these assets to the
TIP™ factors. However, as our qualitative analysis
highlights, there will be opportunities in these
assets in some regions (particularly in the EU and
in emerging-markets areas such as China/East Asia)
that would be beneficial in this scenario.

38
Figure 6
Delayed Action – impact of new portfolio mix on contribution to risk

Equity risk premium


11% Credit risk premium
7% Illiquidity premium
Technology
Policy
16%
54%

12%

Source: Mercer

n Stern Action is a positive scenario for all America, northern Europe and Russia, although as
investments, with a much lower level of risk for Grantham/LSE Vivid Economics pointed out, the
the same level of return compared to the Delayed timing and magnitude of such effects are uncertain.
Action scenario (9% versus 14% risk for the latter).
The results point to an allocation to timberland and n The important caveat to these results is the
agriculture land, as well as a substantial allocation uncertainty around when the market will begin to
to listed sustainable assets. The high allocation to price in the increasing risk of physical damage to the
sustainability-themed assets reflects the fact that environment due to the lack of policy action and/or
under this scenario, liquid assets with a shorter- future carbon costs as they grow. In addition, there
term horizon (such as equities and bonds) will also is much uncertainty in the climate science literature
benefit from the climate-change policy response, about the regional impacts and likely timing of the
as will some illiquid climate-sensitive assets like changes that could take place. These uncertainties
real estate, infrastructure and private equity. This is could increasingly weigh on climate-sensitive assets
because, in contrast to the Delayed Action scenario, and regions in the future. This could affect risk
the Stern Action scenario assumes that the policy premiums on tangible assets such as coastal real
response will be fully anticipated by the market. This estate and infrastructure and agriculture land in
encourages significant transformation not only in flood-prone and/or drought-potential areas.
earlier stage (off-market) investments but also across
the listed equity and debt markets. To conclude, we reiterate that modelling analysis needs
to be interpreted with caution. However, we believe
n Climate Breakdown suggests no significant change that the analysis supports our view that a greater focus
in the portfolio mix within the time horizon of on climate-sensitive assets is likely to be rewarded
this study. Some opportunities may exist within across the climate mitigation scenarios.
asset classes, such as equity, private equity and
infrastructure investments in adaptation measures
(rather than mitigation and technology) such as
flood defence, water management, desalination,
emergency services and disaster relief. Within
agriculture there could also be a rise in crop yields
as temperatures rise in regions such as North

39 Climate Change Scenarios – Implications for Strategic Asset Allocation


40
Asset-class
impacts
The following discussion provides further
evidence to underpin the conclusions
around the investment impacts of climate
change for each asset class.

41 Climate Change Scenarios – Implications for Strategic Asset Allocation


The analysis is qualitative, applying judgement and
interpretation of the climate-change risks and evidence,
as presented in this report. The key assumptions that
underpin these conclusions are as follows:

n The interpretation of the investment impact is


based on an existing portfolio of assets held today,
projecting forward to consider the impact in 2030 for
each scenario in question. The conclusions would be
different if we were to consider building a portfolio
of new assets starting in 2030. We focused on the
impact for existing assets to analyse the current risks
and opportunities that institutional investors need to
manage today, in view of the climate scenarios and
future outcomes.

n We have assumed that all the core asset classes


examined in this study have not integrated climate
change considerations at this present point in time.
This means our analysis shows the range of potential
impacts for portfolios that are “unsustainable”
versus those that are more sustainable,19 which was
considered desirable in view of the aim of this study
to analyse the tail risks associated with climate
change. For institutional investors that have shifted
the underlying nature of their asset exposure to be
more sustainable, it may be appropriate to consider a
sensitivity and direction of impact that is somewhere
between the two.

19 Sustainable assets refers to investments that generate a substantial proportion (typically, more than 25%) of their earnings through sustainable activities. At its
broadest level, sustainable investment seeks to support sustainable economic development, enhance quality of life and safeguard the environment. This includes
sustainable themes such as energy efficiency, low energy transport, renewable energy, bioenergy, carbon capture and storage, smart grid, water supply, usage and
management, waste management, hydro energy, geothermal and biofuel, to name a few.

42
Listed equities

Equities may experience some degree of It is no surprise that the utilities sector is the hardest
transformation under all the mitigation scenarios, hit, followed by basic resources and industrial goods
although we expect most of the transformation should and services. The magnitude of the impact (tonnes of
take place at the sector and/or regional level. The CO2e emissions *future carbon price per tonne of CO2e)
transformation is also likely to happen in response to demonstrates how important the policy outcomes for
policy developments, rather than ahead of them, as the sectors in terms of the future adjustment costs
research indicates that the shorter-term pressures that they will have to absorb. We acknowledge that
that investors and managers of listed companies come some sectors may be more/less able to pass on or
under are likely to constrain large-scale investments absorb this cost than others and that this should also
until the policy framework supports such a shift.20 be considered against future profits for the sectors.
Nevertheless, the scale of the potential cost provides
If the mitigation response is strong, the industries some indication of the level of adjustment that will be
that would be worse off include fossil-fuel industries required across different sectors.
(coal mining, crude oil and gas extraction, petroleum
refining, gas utilities), as well as certain carbon- On the other side of the coin, the winning sectors
intensive primary and manufacturing industries, would include firms operating within low-carbon
including mining, most electric power utilities, and sectors at bottleneck positions in the supply chain.
chemicals. Figure 7 shows the potential cost of the These include the renewable and nuclear power supply
different carbon prices across the scenarios in this chains, carbon capture and storage (CCS), biofuels
report, broken down by sector, based on current and energy efficiency technologies such as smart grid
emissions levels. components, and energy-use auditing methods.

Figure 7
Cost of carbon adjustment by sector
15 Climate Breakdown

110 Stern Action/Regional Divergence

220 Delayed Action

700

600

500
$US billion

400

300

200

100

0
Utilities Basic Construction & Food & Chemicals Travel & Retail Automobiles Personal &
Resources Materials Beverage Leisure & Parts Household Goods

Source: Sector tonnes of CO2e emissions data sourced from Trucost; Mercer calculations based on carbon price across the scenarios, where carbon price is shown as
$ / t CO2e

20 See WWF, Trucost and Mercer Carbon Risks of UK Equity Funds (2009), available at http://www.trucost.com/_uploads/downloads/Carbon_Risks_in_UK_Equity_Funds.
pdf; see also Mercer 2009, Research Identifies Large Scope to Improve Carbon Footprint of Investment Portfolios, available at http://www.mercer.com/summary.
htm?idContent=1351415, accessed 11 January 2011.

43 Climate Change Scenarios – Implications for Strategic Asset Allocation


The countries most dependent on high-emitting As can be seen, the highest-risk countries highlighted
industries at present would therefore experience in red, according to the Brenkert et al study, are
the greatest shift in demand and be relatively worse South Africa (8%), India (8%) and Morocco (0.2%). This
off. Vivid Economics (2009)21 identified the bottom means that around 16% of an emerging-market equity
five countries in terms of sectoral composition to investment potentially faces costs around adaptation
be Australia, the US, Saudi Arabia, South Africa and to climate change. Grantham LSE/Vivid Economics
Canada. The top five countries on the same measures estimate the adaptation and residual damage costs
are China/East Asia, Brazil, South Korea, Mexico and under the worse-case Climate Breakdown scenario
France. would be $71 bn, or 0.7% of the level of GDP in India/
South Asia in 2030, and $56 bn, or 2.1% of the level
Additional factors to consider are not only the of GDP in Sub-Saharan Africa in 2030. Based on this
adjustment to the cost of carbon as a result of policy evidence, we have assigned a higher sensitivity on
and technological developments but also the physical emerging-market equities (moderate) to the TIP™ risk
impact risks that might affect countries to varying factors than for global equities (low).
degrees. In other words, to what extent would a
global equity portfolio be exposed to impact risks At the overall asset-class level, aside from the
over the coming 20 years? And how much more sector and regional differences, the magnitude of
concentrated is that risk for a global equity portfolio the impact in terms of risk/return assumptions is
versus a more focused emerging-market equity expected to be greater for emerging-market equities,
portfolio for which the impact risks tend to be higher? broad sustainability-themed equities and efficiency/
While the timing of these risks is uncertain (as renewables. We also expect the most important
Appendix A illustrates), awareness of where the risks climate factor risks for listed equities would come
might lie over the coming decades would be helpful from technology and policy, although impact risks are
as part of overall risk management. notable for some emerging markets. A few highlights:

Figure 8 shows the weighting of the MCSI EM equities n Technology – This is the key enabling factor for
index as at the end of September 2010 to the most economic transformation due to climate change,
vulnerable countries, based on a study by Brenkert et and while this may have a much greater impact
al (2005). While estimates across studies vary in terms in early stage investments such as private equity
of the potential timing and cost of climate-related and infrastructure, it will also have a knock-on
damage, this is a helpful starting point for investors effect in listed equity as successful technologies are
to consider and monitor the potential source of risk rolled out and emerging companies become more
over time. established. Investment in new technology has a
negative impact on company cash flows initially,
whether through research and development or
Figure 8 through buying proven technology, but corporate
Exposure of EM equity index to climate ‘impact’ risks investment would focus on technologies expected to
Taiwan generate a positive rate of return over the economic
Russia
Mexico cycle. Sustainability-themed and efficiency/
Other
renewables equities, with their clear focus on
Korea this type of investment, would have the greatest
India exposure to this factor.
8%
n Impacts – The sensitivity of global equities to the
South Africa physical impacts of climate change within the time
China
Morocco 8% horizon of this report is expected to be low, although
Brazil 0% specific regions may experience physical changes,
Source: Brenkert M et al (2005), as quoted in Yohe GE et al (2006). “Global usually but not exclusively in emerging-market
Source: Vulnerability to Climate Change”. The Integrated Assessment Journal
6 (3):35–44.

21 This is defined as the structure of the composition of an economy in terms of emission intensive industries and activities. See Table 4 rankings that delineate sectoral
composition rank in Vivid Economics G20 Low Carbon Competitiveness (2009), available at http://www.e3g.org/images/uploads/G20_Low_Carbon_Competitiveness_
Report.pdf.

44
countries. This means that concentrated emerging- Emerging Markets EMEA index is Energy compared
market equity portfolios potentially face higher costs with 10.5% for the All World, 10.9% in the EU, 5% in
related to adaptation to climate change that may Asia Pacific and just 1% in the Far East. The winners
affect corporate earnings in those markets. Sectoral would be the industries supplying technology
differences in vulnerability to climate change may and services to the energy-sector transformation.
be concentrated in a relatively small number of These are usually small sub-sectors of the equity
sectors – such as agriculture, forestry, water and market, and this is the opportunity targeted by
tourism – whose performance depends directly on sustainability-themed funds.
weather conditions.
Table 7 presents the sensitivities of global equities,
n Policy – Climate policy would typically increase emerging-market equities, broad sustainability-themed
operating costs through a higher cost of compliance; equities and renewables to the TIP™ factor risks for
however, in certain sectors it may serve to encourage each scenario. The magnitude (low, moderate, high
investment and innovation. We can expect to see and very high) of the sensitivity to climate change
traditionally carbon-intensive sectors – for example, is presented at the top of the table, with the colour
fossil-fuel industries, primary manufacturing, denoting the direction of the impact for each scenario.
mining, chemicals and electric power utilities – lose
out. Correspondingly, regions with a high proportion
of carbon-intensive sectors would be affected
the most. By way of illustration, 26% of the MSCI

Table 7
Sensitivities of global equities, emerging-market equities, broad sustainability-themed equities and renewables to the TIP™ factor
risks

Asset Global equities Emerging-market Broad sustainability- Renewables


sensitivity equities themed

Sensitivity Low Moderate High Very high

Regional Risk of increased Higher volatility in This is a broadly positive Similar to the broad
Divergence uncertainty and volatility emerging-market equities environment for sustainability-themed
due to regional disparity is likely, where a gap sustainability-themed equities, the very high
on climate policy. will open between those equity, with sporadic sensitivity to the climate
Regional differences emerging-market countries policy encouraging risk factors means that
within major sectors will that have the capacity some industries in some supportive climate policy
become exaggerated, and willingness to grow regions to grow strongly. will attract investment
where carbon-intensive as a low-carbon economy Sustainability-themed in renewable energy to
industries in countries with versus those that are not investments stand to leading regions. The most
carbon constraints will as able or willing to adapt. benefit in the leading supportive policies for
become less competitive Current evidence suggests regions – but those in renewable energy currently
relative to companies in that the emerging the “wrong” regions or include parts of Europe –
countries without carbon economies that are sectors will suffer more particularly Scandinavia,
constraints. Multinational positioned to lead in this than traditional equity France, Germany, Spain
companies may find the scenario include China/East portfolios. Policy and – as well as the UK and
cost of operating across Asia, South Korea, Brazil, technology will be the Brazil.
borders increasing due to Mexico, South Africa and dominant drivers of new
a higher cost of complying India/South Asia. opportunities, driven by
with different national cost/efficiency savings as
policies. well as the expectation of
further policy advances.

45 Climate Change Scenarios – Implications for Strategic Asset Allocation


Table 7
Sensitivities of global equities, emerging-market equities, broad sustainability-themed equities and renewables to the TIP™ factor
risks (cont’d)

Asset Global equities Emerging-market Broad sustainability- Renewables


sensitivity equities themed

Sensitivity Low Moderate High Very high

Delayed Higher volatility is likely to Volatility increases for Sustainability investments The policy turnaround
Action negatively impact global some emerging-market perform strongly following will likely lead to
equities, as the climate equities, notably those the announcement of the outperformance of this
policy turnaround is not that continue to operate policy measures, with a sector compared to the
fully anticipated. Carbon- as BAU for the coming 10 more muted performance other types of listed
intensive industries that years and fail to prepare in the preceding period. equity. Policy measures
benefit from the policy for the dramatic policy Significant potential for will directly benefit the
delay over the next 10 U-turn. The fortunes of the outperformance of the companies in this universe,
years will be penalised, emerging economies will theme versus a traditional boosting returns and
particularly if they have diverge when faced with a global equity or emerging- encouraging further
invested in long-term high cost of carbon (such market portfolio. investments.
infrastructure that as Russia, parts of eastern
becomes redundant. Europe and China/East
Asia versus Brazil, Mexico
and South Korea).

Stern Action A period of positive A supportive environment This is a favourable Renewable energy-focused
transformation due to for emerging markets, scenario for sustainability- companies will be major
supportive and transparent with some countries themed equities, with beneficiaries under
policy. Some carbon- also receiving significant supportive policy and this scenario. With the
intensive industries shrink adaptation transfers from technology flows. The highest sensitivity to the
or disappear while others developed markets. Mercer upside will potentially be climate risk factors, they
face increased costs of research shows that most greater than for traditional will benefit more than
mitigation or pollution investors are structurally listed equity funds and global equities, emerging-
penalties. These include underweight emerging- emerging-market equities. market equities or broad
agriculture and forestry market equities, and Over the longer term, sustainability equities.
as well as the energy, hence, supportive climate the sustainability-leading Companies that focus
extraction and chemical policy is likely to further companies will gradually their revenue-generating
industries. The cost of increase the attractiveness be subsumed into the activities on renewables
capital for companies in of emerging markets. core listed equity indices, will outperform, spurring
these sectors will increase. The lower risk associated making it more difficult R&D investment in new
Investment in technology- with the physical impacts to distinguish between technologies, smart-
development companies of climate change under sustainability-themed grid systems, nuclear,
and those that provide this scenario may further equity portfolios and reforestation, electric
goods and services to the enable emerging-market mainstream global equity vehicles hybrid plug-in,
energy sector will expand. companies to benefit from portfolios. solar, biomass and Carbon
the expected growth and Captive Storage.
social development.

46
Table 7
Sensitivities of global equities, emerging-market equities, broad sustainability-themed equities and renewables to the TIP™ factor
risks (cont’d)

Asset Global equities Emerging-market Broad sustainability- Renewables


sensitivity equities themed

Sensitivity Low Moderate High Very high

Climate The evidence points to The absence of This scenario is negative This scenario is negative
Breakdown physical impacts not investment in low energy for sustainability- for investments in the
being a major cost for the infrastructure solutions themed equities, as the renewable energy theme,
markets to absorb at the could thwart China’s absence of policy action as the policy inaction will
aggregate level within the ability to sustain economic will limit returns and cap investment returns
timeframe of this study; growth, with increased future investments in and flows to around
however, there may be an pressure on resources from technologies to cost- current levels, with
impact if equity markets population growth and savings areas. Some investments in efficiency/
price in the expected rising living standards. sectors, such as energy renewables due to
future degradation. Some emerging-market efficiency, remain resilient efficiency/cost savings
Carbon-intensive industries countries will also while in the more climate being the most resilient.
will experience higher experience severe physical policy sensitive sectors,
costs than less-intensive impacts. For example, such as R&D, the pricing
industries, but not to the Grantham LSE/Vivid of renewable assets and
extent that they would Economics estimate the technologies such as CCS
under mitigation policy adaptation and residual will suffer more.
scenarios. damage costs to be $71
bn, or 0.7% of the level
of GDP in India/South
Asia in 2030, and $56
bn, or 2.1% of the level
of GDP in Sub-Saharan
Africa in 2030. Within
the MSCI EMEA index,
the weightings of the
most vulnerable countries
equate to 16% that
potentially face costs
around adaptation to
climate change.
Source: Mercer drawing from various sources, as referenced

47 Climate Change Scenarios – Implications for Strategic Asset Allocation


Fixed income

The impact of the climate scenarios on fixed income debt concentrated allocation potentially faces high
portfolios will vary depending on the type of the costs around adaptation to climate change. Grantham
fixed income asset. We expect that the impact on LSE/Vivid Economics estimate the adaptation and
government bonds is likely to be relatively low, as residual damage costs under the worse-case Climate
government bonds have a comparably high sensitivity Breakdown scenario to be $56 bn, or 2.1% of the level
to sources of risk related to macroeconomic conditions of GDP in Sub-Saharan Africa in 2030. Based on this
rather than climate change factors, the former of evidence, we have assigned a higher sensitivity on
which are only expected to change under the Delayed emerging-market debt (moderate) to the TIP™ risk
Action scenario. factors than for global bonds (low).

For investment-grade credit, the impact should be


broadly similar to that of global equities, with some Figure 9
substantial regional and sector shifts taking place. Exposure of EM bond index to climate ‘impact’ risks
Mexico
The impact would be most pronounced for emerging Other South Africa
debt and a new fixed income vehicle commonly Turkey 12%
referred to as “green bonds” – a government,
Brazil
development bank or supra-national issued instrument
designed to raise finances for expenditure on climate- Thailand
change mitigation and adaptation. For example, the Hungary
European Investment Bank has issued some Climate Russia
Malaysia
Awareness Bonds in recent years, with the proceeds Poland
being used for projects in renewable energy and energy
Source: Brenkert M et al, as quoted in Yohe G et al (2006). “Global Source:
efficiency. The World Bank has issued a series of green
Vulnerability to Climate Change”. The Integrated Assessment Journal
bonds since 2007 for similar purposes, and more 6 (3):35–44.
recently the IFC issued its first Green Bond to raise
money for investing exclusively in renewable energy,
energy efficiency and other climate-friendly projects On balance, we expect that the most important source
in developing countries. The US Treasury also issued of climate change risk for fixed income assets would
Clean Renewable Energy Bonds in its 2009 budget.22 come from climate policy that will alter the demand/
supply balance of bonds and physical impacts, more so
An additional factor to consider is the physical impact than technology, as will be explained more fully below.
risks that might affect issuing countries to varying
degrees and how much these risks could affect a fixed n Technology – Government bonds and emerging-
income portfolio. Figure 9 shows the weighting of the market debt have a low sensitivity to technology,
JP Morgan GBI EM Bond index as at end September as the private sector is more likely than
2010 to the most vulnerable countries, based on a governments to finance such investments via
study by Brenkert et al (2005). While estimates across debt issuance. However, investment-grade credit
studies vary in terms of the potential timing and cost would experience a degree of transformation under
of climate-related damage, this is a helpful starting the climate scenarios, particularly in the carbon
point for investors to consider and monitor the sensitivity sectors and those that play a key role
potential source of risk over time. in the financing of research and development and
commercialisation of new low-carbon enabling
Figure 9 highlights in red the country identified by the technology. The extent to which green bonds are
Brenkert et al study as being the most vulnerable to the used to finance investment in emerging markets and
physical effects of climate change. As can be seen, the developed economies would also affect the demand/
only high-risk country in terms of climate vulnerability supply balance for bonds and the availability of
within this index is South Africa (highlighted in red capital for investment in the development and
at 12%). This means that 12% of an emerging-market deployment of technology.

22 Further details on green bonds are available at http://climatebonds.net/.

48
n Impacts – Certain sectors and regions are vulnerable be implemented via a carbon tax that would
to extreme weather events, and the risks increase increase the tax receipts received by governments. If
beyond 2050. Climate-vulnerable regions would carbon pricing was implemented via cap-and-trade
be negatively affected by the impact of climate schemes, then the solution would be market based
change that could escalate over time, with emerging and, depending on the design, potentially bring
markets being the most sensitive. For this reason, lower implications for public finances. Emerging
we have assigned a higher sensitivity of emerging debt is likely to be more sensitive to climate policy
market to impact risks, as it may directly affect developments than sovereign debt or investment-
demand for, and supply of, emerging debt and green grade credit. This is because some emerging
bonds to finance adaptation measures in vulnerable economies have a higher sensitivity to the climate
countries (such as India/South Asia and Africa). policy framework, particularly around decisions
related to adaptation payments from developed
n Policy – Under some scenarios, climate policy to developing economies, which, in turn, may be
is likely to lead to increased public spending on financed via green bonds.
energy infrastructure and on other public goods
(for example, the provision of information on Table 8 presents the sensitivities of government bonds,
energy-saving measures) and the promotion of emerging debt, investment-grade and credit to the
energy efficiency in the public sector. The extent to TIP™ factor risks for each scenario. The magnitude
which this investment is financed by private-sector (low, moderate, high and very high) of the sensitivity of
issuance rather than by the public sector would the asset to the climate factors is presented at the top
be important for gauging future government bond of the table, with the colour denoting the direction of
issuance needs. Carbon pricing could, for example, the impact.

Table 8
Sensitivities of government bonds, emerging debt, investment grade and credit to the TIP™ factor risks

Asset Government bonds Emerging market debt Investment-grade credit

TIP™ sensitivity Low Moderate Low

Regional Governments with a proactive The market and/or credit rating Credit rating agencies may begin
Divergence approach to climate policy could agencies may attach a higher risk to factor in future climate risks,
issue more debt to finance premium to some emerging-debt which would exaggerate the
expenditure on programs to shift issuers that are lagging in terms of differences between leading and
to a low-carbon economy. These climate policy response (e.g. Russia) laggard companies in terms of the
may be hypothecated financing and/or are more vulnerable to the sectors they operate in, including
instruments (such as green bonds). physical impacts of climate change fossil-fuel industries (coal mining,
Countries (e.g. Russia, Canada, (such as India/South Asia, Africa and crude oil and gas extraction,
the US, Australia) that are heavily parts of China/East Asia). petroleum refining, gas utilities)
dependent on high-emitting sectors and carbon-intensive primary and
that lag in terms of climate change manufacturing industries, including
policy may attract a higher country mining and chemicals.
risk premium.

49 Climate Change Scenarios – Implications for Strategic Asset Allocation


Table 8
Sensitivities of government bonds, emerging debt, investment grade and credit to the TIP™ factor risks (cont’d)

Asset Government bonds Emerging market debt Investment-grade credit

TIP™ sensitivity Low Moderate Low

Delayed Action After some delay, subsidies and/or Higher risk aversion in the market Companies focused on low-carbon
taxation promote rapid deployment due to dramatic policy measures deployment attract a premium;
driven by the public sector. The could lead to a higher risk premium companies with high-carbon
possible introduction of a carbon attached to emerging debt initially sensitivity suffer cost imposition.
tax could be inflationary and after the policy U-turn. The policy An unexpected introduction of
negative for an existing bond measures are dramatic but focused a high cost of carbon could be
portfolio, although bonds will likely on local measures rather than inflationary. This could place
benefit from safe-haven status as supportive of emerging markets upward pressure on the cost
risk appetite declines in the initial and adaptation needs, increasing of financing, particularly for
aftermath of the policy measures. the role and importance of companies in carbon-sensitive
development banks (such as green sectors of the economy.
bond issuance).

Stern Action Increased bond issuance likely to Supportive environment for Coordinated policy and technology
help finance public spending on emerging markets from climate development provides new
energy infrastructure and on other policy, leading to a decline in opportunities for some corporate
public goods related to climate- emerging-market debt risk issuers to evolve, as well as
change policies and the promotion premiums. Adaptation transfers sufficient time for negatively
of energy efficiency (possibly via from developed nations to those impacted sectors to adapt and
green bonds). The scenario assumes developing nations that are transform to a low-carbon
the private and public sectors vulnerable to climate change reduce economy.
will share the adjustment costs, future impact risks.
and hence, the impact of budget
deficits (and bond issuance) is
expected to be neutral.

Climate This scenario is broadly neutral for Climate-vulnerable regions may be As for global equities, the evidence
Breakdown government bonds, with a risk of negatively affected by the impact points to physical impacts not
rising financing risks for developed of climate change that will escalate being a major cost for the markets
countries as future adaptation over time. While the regions most to absorb at the aggregate level
support for vulnerable regions vulnerable are not major issuers within the timeframe of this report;
increases. of debt held in emerging-market however, there may be an impact
debt portfolios, there is likely to if credit markets and/or rating
be a re-pricing of risk towards agencies start to build expected
emerging-market assets. Currently, future degradation into risk
South Africa is the most vulnerable premiums.
country that features in the JPM EM
Bond Index, with a 12% weighting.
This could increase as other
emerging-market countries develop
their local bond markets.
Source: Mercer, drawing from various sources, as referenced

50
Private equity

Private equity assets will be a core part of the more so than traditional listed equity investors. The
transformation to a low-carbon economy across all structure of the traditional private equity portfolio is
the mitigation scenarios, as investment in low-carbon likely to make a gradual transition to include more
technology would require a high degree of private energy-related low-carbon investments that may
sector financing due to the lack of public sector be more in the venture capital end of the spectrum,
finances with structurally high government deficits, although leveraged buyout (LBO) activity is also
as well as the short-term constraints that are likely to expected to increase. This conclusion is reinforced
limit the extent to which the public market is willing to by a private equity study by Bain & Co. (2010)24 that
invest. predicted the strongest growth area expected within
the private equity market would come from the energy
On balance, we expect that the most important source sector.
of climate change risk for private equity assets will
come from climate policy and technology (for venture The opportunities within private equity are highest
capital funds), rather than from impacts, as will be under the Regional Divergence and Stern Action
explained more fully below. scenarios, driven by policy and technology. In terms
of the sectors within private equity, the greatest
n Technology – Venture capital is highly sensitive to opportunities are likely to be in the renewable energy,
technology, as one would expect, given that private CCS, smart grid, electric cars and battery charging
equity could play a key role in the research and and replacement points, water/waste including
development and commercialisation of low-carbon underground reservoirs (water storage), increased
enabling technology. This is particularly pronounced membrane treatment, biogas and desalination plants,
for renewable-energy venture capital. and R&D across all sectors for new breakthrough
technologies and deployment of technologies in the
n Impacts – This is a low source of risk for private shift to a low-carbon economy.
equity out to 2050, although this would vary by
business type and location, which would need to be The risks are highest for private equity under the
taken into account at the asset placement level. Delayed Action and (to a lesser extent) the Climate
Breakdown scenarios due to an increase in uncertainty
n Policy – There is a high degree of sensitivity across surrounding policy and delay in investment in
the board to changes in climate policy that may technology solutions. Under Delayed Action it is
be introduced via legislation or other market- also assumed that a largely unexpected increase in
based mechanisms. Policy changes will help enable the cost of carbon hits the high-carbon sectors of
and encourage deployment of new technologies an existing private equity portfolio quite hard. The
via private equity through creating appropriate investments most at risk would include investments in
incentives and pricing signals. Indeed, many private the industrials sector, which, together with consumer
equity funds with a cleantech focus to date have products and health care, account for more than 60%
tended to focus on the later-stage deployment and of the private equity market (Bain & Co 2010:22).
commercialisation of proven technologies rather
than on taking technology development risk. In addition to the cost of carbon risk for companies
within a private equity portfolio, the risk of a flood of
Private equity investors tend to have a three-to-five- capital that leads to a bubble in valuations is also a
year horizon when evaluating direct investment risk under the Stern Action and Regional Divergence
opportunities23; hence, the investments that will scenarios that would need to be taken into account.
proliferate are likely to be those that attract the Under these scenarios there is also an increased risk
greatest potential return within that horizon. This that technology advancements move quickly and
would allow private-equity investors to be slightly supersede existing investments.
ahead of government policy and to take some risks
around technology in terms of deployment – at least

23 This is distinct from the horizon that an institutional investor might have when allocating to a private equity fund, which typically has a much longer horizon given
high upfront costs and the j-curve effect, where assets take some time to place and, hence, yield a return.
24 Bain & Co (2010).

51 Climate Change Scenarios – Implications for Strategic Asset Allocation


Table 9 presents the sensitivities of private equity LBO, asset to the TIP™ risk factors is presented at the top of
venture capital and renewable energy to the TIP™ the table, with the colour denoting the direction of the
factor risks for each scenario. The magnitude (low, impact.
moderate, high and very high) of the sensitivity of the

Table 9
Sensitivities of private equity LBO, venture capital and renewable energy to the TIP™ factor risks for each scenario

Asset Private equity LBO Private equity venture capital Private equity renewable energy

Sensitivity Moderate High Very high

Regional The transformation that will As for LBO, with the key difference Renewable investments may be
Divergence take place in some regions that being a very high opportunity related highly sensitive to the climate policy
implement climate policy and invest to technology in some regions. variability by countries. The regions
in technology will be significant, with The countries with the highest with the most supportive policies for
LBO activity expected to increase expenditure on low-carbon solutions renewable energy and the deepest
as the economies in those regions and the deepest venture capital investment markets based on the
shift from high- to low-carbon markets at present include the EU, current policy environment and
industries. The economies likely to the US, China/East Asia, Japan, the clean energy markets include parts
experience the greatest shift in this UK and parts of Latam (Brazil and of Europe – particularly Scandinavia,
transformation are those that are Mexico). France, Germany and Spain – and
high-emitting nations but implement the UK, China/East Asia, states
policy measures – at present, this within the US, Brazil, India/South
could represent the EU, China/East Asia and Japan.
Asia, the UK, states within the US,
and Japan.

Delayed Subsidies and/or taxation (hard On the basis that policy changes The policy U-turn may lead to
Action regulations) promote rapid are not fully anticipated, private strong performance of renewable
technology deployment. Funds equity venture capital assets that assets after the measures are
focused on low-carbon deployment are dominated by high-carbon implemented. Due to late action the
attract premium, while existing investments would face increased policy response will focus more on
private equity funds with high- unexpected costs in response to a deployment of existing technology.
carbon sensitivity suffer cost dramatic policy shift. If a private The main “proven” technologies
imposition. An increase in equity portfolio has a higher degree include wind, solar, sugar-based
bankruptcies is likely for a period of low-carbon investments as the ethanol, and cellulosic and next-
following policies and a reduction in starting point, then the assets would generation biofuels.
LBO activity. be more resilient.

Stern Balance between R&D and As for LBO, but the activity is A positive scenario, as these assets
Action deployment is likely due to expected to focus on identifying new should play a key role in the early
supportive policy environment. opportunities for both development stages of the R&D development
Opportunities extend to new and and deployment of new and deployment (including wind,
existing funds to capture low- technologies. Further technology solar, CCS and geothermal). Over
carbon transformation investments, risk may be taken into venture the long term, the “mainstreaming”
with policy clarity and consistency capital funds than is currently the of renewable energy may lead
reducing uncertainty. An increase case, encouraged by the supportive to a similar risk/return profile to
in LBO activity is likely as a period policy framework that makes such traditional venture capital private
of creative destruction unfolds as investments economically viable. equity funds, as there would be less
companies in low-energy sectors opportunity for specialist portfolio
outperform high-carbon or energy- managers to have an informational
intensive businesses. advantage over their generalist
peers.

52
Table 9
Sensitivities of private equity LBO, venture capital and renewable energy to the TIP™ factor risks for each scenario (cont’d)

Asset Private equity LBO Private equity venture capital Private equity renewable energy

Sensitivity Moderate High Very high

Climate Neutral overall for LBO activity in the Same as for LBO, but higher This is a negative scenario for
Breakdown timeframe of this study, although sensitivity to “no mitigation” venture capital renewable energy,
higher physical impact risks will policy will reflect the possible particularly for existing private equity
need to be priced into certain types impact on existing private equity asset valuations that have likely
of assets. New opportunities will asset valuations for low-carbon priced supportive policy action within
proliferate in adapting to climate investments that have been priced in this horizon (out to 2030).
change in the absence of mitigation, policy action (i.e. they could lead to
such as flood defence, water a downward re-pricing of cleantech
management and desalination. The assets held in a portfolio).
nature of investments in underlying
companies in terms of type of
business and physical location
increases in importance as part of
the due diligence process.
Source: Mercer, drawing from various sources, as referenced

53 Climate Change Scenarios – Implications for Strategic Asset Allocation


Infrastructure

Infrastructure assets will be a core part of the any new building infrastructure or improvements.
adaptation and mitigation efforts of governments Some of the long-term risks include a flood risk for
around the world. As such, the decarbonisation of new/ assets in coastal areas, damage caused by knock-on
existing assets as well as adaptation to climate change effects of heat waves and damage caused by storms
through the replacement of assets or construction of (wind, rain, snow). This could lead to interruptions
new assets will be important drivers behind long-term in electricity and water supply, disruptions to road
infrastructure investment trends internationally. and rail networks, softer road surfaces, restrictions
on water usage, power station inoperability (for
The long-term nature of infrastructure investments, example, nuclear stations in France during the heat
with a 10+ year horizon when evaluating the wave during 2010) and blackouts. The evidence
attractiveness of an asset, means that future policy suggests that these risks may vary by investment
changes and technology advances, as well as type and location, which would need to be taken
physical impacts due to climate change, are more into account as part of the evaluation of the
likely to be taken into account in the evaluation investment opportunity (be it via a fund manager or
process. Moreover, we expect that the nature of through direct investments).
a traditional infrastructure portfolio would likely
transition over time to embed more of the new n Policy – Given the public/private partnership
opportunities available and to move away from nature of many infrastructure investments and the
high energy intensive assets. This could entail, for socioeconomic needs that infrastructure assets fulfil,
example, a shift in the nature of an infrastructure policy and regulatory changes related to climate
portfolio to a higher allocation of non-core assets, change should be important drivers of infrastructure
such as development projects and emerging market returns under all the mitigation scenarios. Such
investments, than is currently the case. policies might take place at the global or national
level and will likely be implemented using a variety
We expect that the most important source of climate- of approaches for the design of new, and retrofit
change risk for infrastructure assets will come from of existing, infrastructure with respect to climate-
a combination of climate policy and technology, with change mitigation and adaptation. For example,
impact risks being limited to a more narrow set of this may involve changes to planning applications
assets in particular regions. A few overall observations at regional and local levels, as well as new standards
on the sensitivity of infrastructure to TIP™ factors: of engineering and measures to encourage
integration of solutions into planning processes
n Technology – Technology should play a role in and construction.
enabling the upgrade and adaptation of existing
infrastructure towards low-energy solutions. For Overall, the biggest opportunities within infrastructure
example, energy power distribution could be may be within the Stern Action and Regional
aided by new advances regarding smart grids and Divergence scenarios, driven by both policy and
technologies in renewable energy, nuclear power and technology factors. The location of infrastructure
CCS. Transport transformation could be assisted by assets may be important for investors to consider, as
electrification of rail and motor vehicles and new many of the fast-growing developing economies have
technologies in road surfacing. Improved drainage an opportunity to build new infrastructures that are
and flood-protection measures in airports could take more climate-resilient from the outset, potentially
place. New advances in irrigation, water storage and creating an advantage over developed countries whose
desalination could also have an impact on water. existing assets have high replacement and upgrade
costs that will slow down their rates of investment.
n Impacts – The long-term nature of infrastructure In terms of types of infrastructure assets by sector,
assets and the fact that many are built with a 100+ the greatest opportunities related to climate change
year life span increases the importance of due for institutional investors are likely to be in energy,
diligence and environmental risk assessments in transport and water/waste.25 Within energy, the

25 URS Adapting Energy, Transport and Water Infrastructure to the Long-Term Impacts of Climate Change (2010).

54
areas include renewables, nuclear and CCS, but with long term). This risk is highest under the Stern Action
more focus also on transmissions and distributions and Regional Divergence scenarios, where the rates of
networks, decentralised electricity and heat generation, technological progress are expected to be the highest.
additional fuel capacity storage, electric vehicle The highest impact risks on some infrastructure assets
recharging points, hydrogen and biogas. due to physical changes to the environment would
be under the Delayed Action and Climate Breakdown
The opportunities within transport include road, scenarios, where late or no action on policy increases
rail and bridge replacements; sustainable drainage the prospect of higher environmental damage in the
systems; electrification of rail and overhead electrical future (and therefore higher risk premiums for some
lines; electric cars and battery charging and infrastructure assets may be required).
replacement points; road surfacing; improved drainage
and flood-protection measures; and larger berths and Policy or regulatory risks would be highest under
improved port design. Regional Divergence, where there may be some
uncertainty about the variation on the direction of
In water/waste, opportunities could include policy across regions and hence the long-term path of
underground reservoirs (water storage), increased investment in infrastructure projects. Delayed Action
membrane treatment, biogas and desalination plants. would also be disruptive to the extent that the abrupt
policy changes are not anticipated in project planning.
Just as these sectors pose the greatest opportunity,
they also present the greatest potential risk under the Table 11 (on page 56) presents the sensitivities of
“no mitigation” scenario of Climate Breakdown. In infrastructure core unlisted and renewables unlisted to
brief, technology advancements could reduce the value the TIP™ factor risks for each scenario. The magnitude
of some infrastructure assets that are less advanced (low, moderate, high and very high) of the sensitivity
or unable to utilise the improvements or, in the most of the asset to the TIP™ risk factors is presented at the
extreme case, they could render some infrastructures top of the table, with the colour denoting the direction
redundant (coal power stations perhaps, in the very of the impact.

Table 10
Infrastructure types by sector

Economic Social

Transport Energy Water/Waste Communication

Roads, bridges and Pipelines Water distribution Cable networks Health care (hospital,
tunnels etc.)
Fuel processing, Waste collection Satellites
Railway networks storage and transport Education
Water supply Transmission/
Airports Contracted power broadcasting Penitentiary
Water treatment infrastructures
generation and
Ports
pollution control Waste treatment and
Ferries disposal
Energy distribution
systems
Source: Adapted from a combination of CDC (2010) report and Adapting Energy, Transport and Water Infrastructure to the Long-term Impacts of Climate Change, URS,
January 2010. A report commissioned by the UK government’s cross-departmental infrastructure and adaptation project

55 Climate Change Scenarios – Implications for Strategic Asset Allocation


Table 11
Sensitivities of infrastructure core unlisted and renewables unlisted to the TIP™ factor risks for each scenario

Asset Infrastructure, core unlisted Infrastructure, renewables unlisted

TIP™ sensitivity High Very High


Regional This scenario produces some new opportunities The high sensitivity to the climate factors means that
Divergence for infrastructure, but it increases volatility due investments may be highly sensitive to the climate
to political and regulatory uncertainty over which policy variation by countries. The North American
regions will lead and lag. Replacement of existing market is focusing on smart-grid and technology
infrastructure will generally be highest in developed solutions to improve efficiency of delivery more than
economies where stock is old, unsuitable for climate adaptation of infrastructure assets. Electrification
change and unsustainable from an energy-efficiency of vehicles and recharge solutions may also attract
perspective. Investment in new infrastructure that is investments. The UK and Europe are leading the
geared towards low carbon and adaptation to climate development and deployment of many renewables
change will be stronger in the fast-growing developing and decentralisation of electricity generation. The
economies, including China/East Asia, Brazil, Mexico regions with the greatest need for water storage and
and South Korea. desalination include those facing water shortages as
population growth and changing climate conditions
reduce availability. The studies suggest this will be
in MENA, Central America, the southern portion of
the US, southern countries in Africa, and southern
Australia.

Delayed Action This scenario brings a mixed profile for infrastructure, The higher sensitivity of renewable energy
as policy changes occur late and are costly for existing infrastructure assets to the climate risk factors
assets that have not fully anticipated the adjustments means that the policy U-turn will lead to strong
required. It could produce new opportunities for performance after the measures are implemented.
assets that price in additional costs and risks as Due to late action the policy response will focus more
compensation. A possible inflation rise owing to a on deployment of existing technology rather than on
carbon price shock is generally supportive in some R&D. The main “proven” technologies include wind,
infrastructure assets, which could offset cost pressure solar, sugar-based ethanol, and cellulosic and next-
from policy. Overall volatility increases to reflect generation biofuels. As well as the energy sector, it
unexpected adjustment costs. may also bring investments in transport efficiency
infrastructure and water/waste management that focus
on cost/efficiency savings.

Stern Action This scenario brings attractive opportunities for This is a positive scenario, as renewable infrastructure
investors, with policy clarity and investment reaching investments will play a key role in the early stages
high levels in low-carbon infrastructure assets across of both R&D and deployment. Opportunities include
transport, energy and water/waste globally. Over the renewable energy; road, rail and bridge replacements;
long term, the “mainstreaming” of renewable energy sustainable drainage systems; electrification of rail
may actually lead to some merging between the and overhead electrical lines; electric cars and battery
composition of a core unlisted infrastructure asset and charging and replacement points; road surfacing;
low-carbon/low energy assets. improved drainage and flood-protection measures;
and improved port design. In terms of water/waste,
opportunities include underground reservoirs (water
storage), increased membrane treatment, biogas and
desalination plants.

56
Table 11
Sensitivities of infrastructure core unlisted and renewables unlisted to the TIP™ factor risks for each scenario (cont’d)

Asset Infrastructure, core unlisted Infrastructure, renewables unlisted

TIP™ sensitivity High Very High


Climate Coastal infrastructure will likely attract a premium This is a negative scenario for sustainable infrastructure
Breakdown to reflect risk of damage from rising sea levels assets, particularly for planned or existing investments
and increased severity of storms/extreme weather. that have been built on the assumption of future
For energy assets, climate change may lead to supportive policy action.
transformer thresholds being exceeded and energy-
generation efficiencies could be negatively affected;
some reinforcement may be required of energy
transportation and distribution systems. Low-lying
areas most prone to flood risk where water availability
risk is highest include those in MENA, Central America,
the southern part of the US, southern countries in
Africa, and southern Australia. Flood and drought risks
need to be built into the design of new infrastructure
and replacement of existing stock.
Source: Mercer, drawing from various sources, as referenced

57 Climate Change Scenarios – Implications for Strategic Asset Allocation


Real estate

As for infrastructure, the long horizon of real estate savings occur when buildings are designed from
and the “real” nature of investments, in the sense that scratch with energy efficiency in mind (which will
it constitutes a tangible asset, increase the importance be possible in many emerging markets). However,
of climate-change risk factors as an extension of due to the long lives of buildings27 and the large
evaluating the risk/return profile of property assets global stock of inefficient buildings, the largest
over the long term. Climate policy is the key driving carbon-saving potential over the next few decades
force, but advances in technology with regard to is actually from retrofitting (in particular, installing
energy efficiency are equally as important, as will be better insulation to reduce heating and cooling
evaluating the risks around potential damage due to needs), not from new buildings.
climate-related risks. To varying degrees, real estate
portfolios are starting to transition towards a greener n Impacts – The physical impacts of climate change
portfolio as new environmental regulations and are expected to be relatively small between now
building efficiency measures increase the standards of and 2050. However, the future risk of extreme events
property assets. and weather changes would be part of how the
market evaluates risk under some scenarios. The
There are a number of different types of vehicles main changes that are anticipated are in demand for
through which investors can access real estate heating and cooling as well as for protection against
investments; this paper focuses on unlisted (direct) intense precipitation and flooding (both coastal
core assets. As Table 12 (on page 59) illustrates, storm surges and fluvial).
climate change has the potential to affect real estate
income return through changes in operating costs and n Policy – Given that the building sector contributes
occupancy rates. In addition, capital growth may be to 30% of global greenhouse gas (GHG) emissions
affected through changes in depreciation and expected and is a sector in which emissions can be reduced
rental growth. relatively cheaply with proven technologies,28 it may
be a key target for policymakers under mitigation
We expect that the most important source of climate- scenarios. An estimate from McKinsey shows that
change risk for real estate assets will therefore come emission-abatement potential in the building sector
from a combination of climate policy and technology, could lower emissions in 2030 from 12.6 Gt CO2
with impact risks being limited to a more narrow per year to 9.1 Gt CO2 per year. Furthermore, the
set of properties in particular regions. A few overall Intergovernmental Panel on Climate Change (IPCC)
observations on the sensitivity of real estate to reported that net-cost additions to achieve stabilised
TIP™ factors: CO2 levels by 2050 may be 7% of total building costs
worldwide.29 Without a global response to mitigation
n Technology – Most technologies available in the (that is, the Climate Breakdown scenario), emissions
real estate sector have already been proven in the from buildings could more than double in the next
market with predictable performance and cost.26 20 years.30
These focus on energy efficiency and demand
reduction. The largest and most cost-effective

26 McKinsey and Co. (2009), p. 105.


27 McKinsey estimates that the overall lifespan of buildings is approximately 35–70 years, with the average being as long as 60–70 years in developed countries.
28 The most basic technological options for mitigating existing buildings’ emissions include more efficient heating, ventilation and air conditioning (HVAC) systems;
reduced electricity consumed by lighting and appliances; solar thermal water heaters; heat pumps; and insulation and double glazing.
29 World Business Council for Sustainable Development Energy Efficiency in Buildings: Transforming the Market (2009), p. 1.
30 UNEP Sustainable Buildings and Climate Initiative, Buildings and Climate Change: Summary for Decision Makers (2009), p. 3.

58
Table 12
Real estate and climate change impacts

Climate change transmission mechanism Return drivers

Increase in cost base due to regulatory requirements leading to Operating Tax, maintenance, replacement,
higher upfront cost of building materials, fittings, etc. costs depreciation

Reduced utility costs, tax and depreciation due to efficient water Water and energy costs, tax, depreciation
and energy technologies

Higher utility costs and insecurity, driven by higher water and Water and energy costs, maintenance,
energy insecurity insurance

Increased/reduced utility costs due to higher requirement for Water and energy costs, tax, maintenance,
cooling – less heating in high latitudes depreciation, insurance

Risk of flooding and extreme weather conditions leading to Occupancy Location


location discounts/premiums in rental incomes rates

Increased occupancy rates due to reduced utility costs from Location and expected rental growth
efficient water and energy technologies (i.e. increased demand for
“green buildings”)

More tourism in mid- to high-latitude regions and decreased Travel demand (countries that rely more
tourism at low latitudes heavily on tourism will be prepared to
invest more into the infrastructure of the
area to promote construction and tourism)

Changes in access to logistics will affect occupancy rates Logistics


Source: Mercer, adapted from various sources

According to RREEF research,31 the leading markets in Overall, the opportunities within real estate are highest
terms of sustainable building include western Europe, under the Stern Action and Regional Divergence
Australia, Canada and Japan. Sustainable practices are scenarios, driven by both policy and technology factors.
still lagging in faster-growing emerging economies,
although these regional variations are expected
to change over time as global flows of capital and
technology intensify. Government regulation would
continue to be a significant driver for energy-efficient
low-carbon buildings in developed and emerging
countries through setting minimum standards for
new construction through building codes (such as in
the US and Europe), efficiency of existing buildings
(India/South Asia), transparency regarding efficiency
rating (Japan), and phasing in escalating sustainability
standards for all residential and commercial buildings
(California).32

31 RREEF Research Globalization and Global Trends in Green Real Estate Investment (2008).
32 ibid.

59 Climate Change Scenarios – Implications for Strategic Asset Allocation


Table 13
Real estate opportunities

Theme Regions

Energy Efficiency in buildings and appliances is the area where most opportunities exist across all regions. However,
and water the greatest potential for low-cost mitigation by 2030 is in electricity savings in non-Organisation for
efficiency Economic Co-operation and Development (OECD) countries from more efficient cooling systems and
appliances. In China/East Asia, measures to promote the uptake of more efficient air conditioning are
significant. Within the OECD, opportunities are primarily in more efficient heating and cooling systems and
appliances from retrofitting, rather than from new build (in particular, installing better insulation to reduce
heating and cooling needs). Out of the OECD countries, the US holds the greatest opportunities for both
new low-carbon construction and retrofits based on projected green construction volume – followed by the
UK, Japan and Germany.33
In terms of the types of real estate, the most rapid transformation is expected in buildings where
considerable energy savings can be made, such as high-rise office buildings, high-profile uses such as retail
centres, and urban in-fill sites.34

Heat pumps According to the Climate Group, heat pumps will be fitted in 50%–70% of buildings in the OECD by 2050.
The US is likely to represent the greatest opportunities in both new construction and retrofits, followed by
the UK, Japan and Germany.

Solar space Policy intervention may reduce capital costs for solar space and water heating, representing notable
and water opportunities in the Southern Hemisphere. Incentives are already established or under way in countries
heating such as Australia, China (where basic models are around 80% cheaper than in Western countries) and
Spain.
Source: Mercer, compiled from various sources as referenced

33 ibid.
34 ibid.

60
The risks for real estate are more apparent under the to the environment that might negatively affect some
Delayed Action and Climate Breakdown scenarios, in real estate assets, depending on their location and/or
both cases driven by an increase in the physical risks carbon preparedness in terms of energy efficiency.

Table 14
Real estate risks

Theme Regions

Lack of carbon An unexpected increase in the cost of carbon could hit real estate across all regions, particularly under
preparedness the Delayed Action scenario. The impact will be larger the lower level of carbon preparedness of the real
estate portfolio.

Sea-level rise Low-lying coastal areas in populated areas such as Bangkok, New Orleans and Shanghai are vulnerable to
and extreme rising sea levels, especially those due to floods and storms. From an investment perspective, the impact of
weather cyclones may be most significant, affecting countries of all income levels, including upper-middle-income
events and high-income levels.

Heating and An increase in heating and cooling demand in the Northern Hemisphere may result in net higher
cooling expenditure on building maintenance to improve insulation and cooling capacity – particularly when
retrofitting buildings.

Precipitation There may be some costs to individual properties to avert storm damage, as well as adaptation costs for
and flooding public works to improve drainage and infrastructure resilience in wetter areas.

Water Water scarcity is expected to be potentially significant in Asia, Africa and Latin America. Risk of water
availability shortages is greatest in Asia (circa 1 billion people would face reduced water supplies and extreme weather
events with a 1–5 degree temperature increase).
Source: Mercer, compiled from various sources as referenced

61 Climate Change Scenarios – Implications for Strategic Asset Allocation


Table 15 presents the sensitivity of core unlisted real is presented at the top of the table, with the colour
estate to the TIP™ factor risks for each scenario. The denoting the direction of the impact (negative, neutral
magnitude (low, moderate, high and very high) of or positive) for each scenario.
the sensitivity of the asset to the TIP™ risk factors

Table 15
Sensitivity of core unlisted real estate to the TIP™ factor risks for each scenario

Asset Real estate, core unlisted

TIP™ sensitivity High


Regional It will be important to consider climate change preparedness and vulnerability at the regional level when
Divergence considering real estate investments. Those regions that are most at risk from the physical impact of climate
change will attract a higher risk premium under the less internationally coordinated emission reduction outcome
as it increases future impact risks. Efficiency in buildings and appliances will be where most opportunities exist.
In China/East Asia, measures to promote the uptake of more efficient air conditioning may present opportunities.
Within the OECD, opportunities will primarily be in more efficient heating and cooling systems and appliances
from retrofitting (in particular, installing better insulation to reduce heating and cooling needs) rather than in new
build.

Delayed Action This scenario primarily brings risks for real estate where policy changes occur late and require short and sharp
adjustment costs. Real estate assets are very sensitive to changes in regulation and will be the target for such
measures and quick action. Investors in real estate assets that are not up to a high standard on energy efficiency
grades will be exposed to risk of obsolescence and high adjustment costs to improve the efficiency of their
buildings after the policy measures have been introduced. Portfolios that have already been “greened” to a high
standard will be more resilient to the policy measures.

Stern Action This scenario brings opportunities to improve energy and water efficiency management. The most rapid
transformation is expected in high-rise office buildings, high-profile uses such as retail centres, and urban in-fill
sites.35 Heat pumps will be fitted in the majority of buildings, with the US leading in new construction and
retrofits, followed by the UK, Japan and Germany. Policy may reduce costs for solar space and water heating.
Incentives exist in Australia, China/East Asia (where basic models are around 80% cheaper than they are in other
countries) and Spain.

Climate This scenario is likely to be negative for real estate assets. Low-lying coastal areas in populated areas, such as
Breakdown Bangkok, New Orleans and Shanghai, are vulnerable to rises in sea level, especially those due to floods and
storms. From an investment perspective, the impact of cyclones may be most significant, affecting countries of all
income levels, including upper-middle-income and high-income levels. An increase in heating and cooling demand
in the Northern Hemisphere may result in net higher expenditure on building maintenance to improve insulation
and cooling capacity – particularly when retrofitting buildings. There may be some costs to individual properties to
avert storm damage as well as adaptation costs for public works to improve drainage and infrastructure resilience
in wetter areas.
Source: Mercer, drawing from various sources as referenced

35 ibid.

62
Commodities

We consider three types of commodities: timberland, n Policy – Climate policy may be an important feature
agriculture land and the carbon market. We focus of how investments in agricultural land, timberland
on these investments since they can be invested in and carbon perform over the coming decade. The
outside a commodities futures basket. This was a key mix of policy options (including but not limited
consideration for our analysis, as it is more difficult to to emission trading schemes and projects) and
ascertain the long-term effect of the climate scenarios the degree to which they promote incentives to
on a broad commodity futures basket, since the impact reduce emissions through sustainable agricultural
may vary quite substantially across different types of techniques, reforestation and avoiding deforestation,
commodities. A few observations on the technology, increased demand for sustainably sourced materials
impacts and policy factors: (such as timberland) and the resultant cost of
carbon will all have an impact on the price and costs
n Technology – Technology would have a limited of production for investments in agricultural land,
impact on agricultural land and timberland, timberland and carbon.
although crop technology would help with
adaptation efforts (for instance, heat-tolerant Agriculture land
and drought-tolerant crops). Biomass electricity
generation and transport biofuels are creating new The uncertainties with regard to climate change
and important markets for forest resources. Under impacts on agriculture in terms of international trade,
mitigation scenarios, this could increase demand markets and investments are very high. Nevertheless,
for sustainably produced forest and manufacturing research suggests that climate change may affect the
residues, as well as for different types of agriculture prices and volumes of goods traded between
for biofuels (although sustainability guidelines are developed and developing countries, particularly
vital around the latter to limit knock-on damage to agricultural raw materials and food, with wider
the environment and food security36). For carbon, macroeconomic consequences.37
since the price is partly determined by the supply
and demand of the carbon emission permits and One of the sources of risk for investors in agricultural
whether the permits can be exchangeable across land relates to how climate policy and the extent
different markets, technology may play a role to the to which emissions from agriculture and land use
extent that the technology enables the reduction could be incorporated into a global mitigation
in emissions but is not the dominant driver of the regime. Emissions from agriculture and land-use
price. change account for about a third of all anthropogenic
greenhouse gas emissions (Smith et al, 2007;
n Impacts – Timberland and agricultural assets/land FAO, 2008). In addition, they raise the prospect of
face a direct risk of damaging physical impacts that extensive low-cost mitigation. According to McKinsey
could substantially reduce the value of investments and Company (2009), roughly a third of the global
(storms, flooding, insect plagues, etc.). Overall, the abatement potential available at a cost of less than $20
impact on agricultural land and timberland is high in 2030, some 9.4 Gt CO2e, lies in reducing emissions
but can vary a lot across regions; in some regions in these sectors, through such options as soil
crop yields will increase, while in others they will management and reduced forest conversion. Therefore,
decrease, possibly making agriculture in certain whether measures to abate these emissions are put
crops no longer feasible. Physical changes in the in place could have a decisive impact on the costs of
environment may also play an important role in agricultural crop production and should be taken into
determining the carbon price over the long term, account by investors.
as the carbon price could increasingly act as a
barometer of climate-event risks.

36 The European Commission released two documents clarifying how member states should implement the biofuel components of the Renewable Energy Directive by
the end of 2010.
37 Ludi et al (2007).

63 Climate Change Scenarios – Implications for Strategic Asset Allocation


Another source of risk for agricultural investments Timberland
relates to the impact of climate change on cash flows
from changes in crop prices. Within the horizon of Timberland investment involves purchasing both
this study, Grantham LSE/Vivid Economics showed plantations and naturally occurring forests in order
that climate change, along with other socioeconomic to harvest the wood on a sustainable basis. Mercer’s
drivers, is expected to place some upward pressure in-house research considers that the major factors
on agricultural prices across all the scenarios. The that drive the performance of timberland as an
scenarios diverge the further the horizon, where the investment include:
“no mitigation” Climate Breakdown scenario expected
to result in a higher price rise out to 2050. As such, n Biological growth – As a tree grows, there will be
across the scenarios the impact of climate change more wood to harvest; hence, its value would
on crop prices is broadly similar, having the effect of increase over time, with the life cycle spanning
driving up prices and, hence, cash flows that some 40–60 years. In addition, the wood from larger
underpin investments. trees is typically used for more valuable products.

But crop prices only tell part of the story for investors. n Timber product prices – The price of timber is
The reliability of crop-yield production and efficiency affected by a number of macroeconomic variables
improvements to agricultural practices, including (including GDP growth, the housing market,
sustainable farming methods, are also important interest rates, etc.) as well as microeconomic
considerations given the rising pressures on water factors (such as government regulation, alternatives
availability, land supply and uncertain weather to wood and rainfall).
patterns. Climate change increases the uncertainties
around crop production and the need to invest in n Land values – Land values are related to local
more sustainable techniques (as will be discussed supply-and-demand conditions and vary from
below); hence, investment in “real” agricultural assets market to market. Land prices, although influenced
as distinct from trading commodity futures would be by timber prices, are typically far less volatile.
affected in different ways.
n Ecology – Additional returns may also be generated
The concept of “sustainable agriculture” seeks to from the voluntary carbon market and projects
minimise environmental damage and to ensure involving forestry and biomass that generate carbon
longer-term productivity through reducing chemical credits (and some fossil fuel switching).39
inputs and energy use in farming systems, promoting
the efficient use of water, the use of complementary Drawing from a study by the World Resource
planting/permaculture, and the development of land Institute,40 one source of uncertainty around climate
for ecosystems services (such as water catchments, change for timberland investments is related to
flood mitigation, biodiversity offsets). Most agricultural physical impacts, where changes in temperature,
assessments of global environmental change have droughts, floods, storms, fires and insect infestations
not focused explicitly on sustainability issues and can reduce forest productivity. A reduction in forest
have neglected the considerable impacts of shifting productivity could have a negative impact on crop
agricultural zones, alterations in commercial fertilizer yields and land values.
and pesticide use, and changes in the demand for
water resources.38 Nevertheless, this is becoming an Climate policy and demand/supply side impacts are
important consideration for investors in agricultural another key consideration for investors. Forests cover
assets and would likely improve the resilience of almost 30% of the world’s land area and deforestation
an allocation to agricultural land across the climate is said to contribute some 18%–25% of total greenhouse
scenarios. gas emissions. The potential reduction in emissions by
avoiding deforestation is said to be as high as 60% of

38 Rosenzweig and Hillel (1995).


39 For further explanation, see http://pdf.wri.org/trees_in_the_greenhouse.pdf.
40 Aulisi et al (2008).

64
potential mitigation by 2030, dominating any increased Carbon
demand for timber-based materials due to more
sustainable building practices (Stern, 2007). Avoiding A number of carbon emission trading schemes have
deforestation would be a broadly positive outcome been established following the adoption of the
for investors in existing timberland assets, as it would Kyoto Protocol in 1997. Carbon emission permits
increase land values, drive up timber product prices, can be categorised into two major types – allowance-
as well as increase investment opportunities around based and project-based. Allowance-based permits
carbon credits. are allocated by regulators under cap-and-trade
schemes, with the major type of allowance-based
On the other hand, such policy measures could reduce permit traded in the EU Emissions Trading Scheme
the appeal of new timber investments depending on (EU ETS). The allowances are called European Union
the extent to which the valuations are driven up by Allowances (EUAs) and are allocated under the
these factors, and also the premium that would be National Allocation Plan of each member country
attached on existing plantations given rising costs of within the EU ETS.42 In addition, there are some
deforestation for new harvests. The detail of the regional voluntary markets such as the Chicago
policy measures around creating the right incentives Climate Exchange and the New South Wales
for investors in new timberland assets would therefore Greenhouse Gas Reduction Scheme in Australia.
be important.
Project-based permits are generated by participation
A report authored by Forum for the Future (2009)41 in certified projects under arrangements such as
discusses the prospects for policy within the reduced the Clean Development Mechanism (CDM) and
emissions from deforestation and degradation (REDD) Joint Implementation (JI). Carbon emission permits
and afforestation, reforestation and sustainable forest generated from these certified projects are called
management (REDD+) schemes. While these schemes Certified Emission Reductions and Emission Reduction
are excluded from Kyoto, under a strong mitigation Units, respectively.
scenario such as Stern Action, we expect they would
feature in the climate policy response and, hence, their This paper focuses primarily on the allowance-
implications would need to be considered by investors. based carbon emission mechanisms, rather than
The Forum for the Future report suggests a number project-based mechanisms. Allowance-based
of areas for policy design of relevance to institutional mechanisms produce a measureable output in terms
investors in timber assets, including creating of the carbon price that is traded on an exchange.
regulations and incentives related to allocation of In addition, the nature and design of project-based
capital to REDD+ projects – such as political influence permits are likely to change significantly over the
and increasing costs associated with governance/ coming five to 10 years as a result of policymaker
compliance – to reduce the risks that may destabilise deliberations in response to feedback on improving
cash flows. these mechanisms. At present, investments will be
constrained by obvious limitations (such as low market
The World Resources Institute (WRI) report concluded liquidity and limited regional focus).43
that many risks and opportunities related to forestry
businesses will vary greatly by a company’s geographic A number of supply-and-demand side factors are
location, position in the value chain, and the influencing investments. The supply side primarily
sustainability of operations, noting that “companies relates to how many carbon permits have been
with experience in sustainable forest management and allocated. The demand side is linked to the volume
supply chains may be better positioned to capitalize of carbon emitted during a period, which in turn is
on new climate change regulations and market forces.” influenced by short- and long-term factors such as
Given the uncertainties that prevail, we believe the weather patterns, relative fuel price, climate policy
same advice holds true for institutional investors in changes, consumption trends and economic growth.
timberland assets.

41 Chapple A. Forest Investment Review (Forum for the Future, DfID and DECC, 2009), available at http://www.forumforthefuture.org/files/Introduction_FIR.pdf.
42 The bulk of exchange trading activity in EUAs is concentrated on the European Climate Exchange (ECX) CFI contracts that are traded on the ICE Futures exchange in
London. ECX accounts for some 80% of EU ETS exchange transactions (including OTC contracts cleared through exchanges), with the balance traded on Nord Pool,
Powernext and the European Energy Exchange. Source: Barclays Capital Global Carbon Index Guide 2008, p. 5.
43 For further discussion on carbon as an investment opportunity, see “Carbon Risk and Carbon Trading: Investment Considerations”, available at www.mercer.com/ri.

65 Climate Change Scenarios – Implications for Strategic Asset Allocation


Unlike other commodities, carbon is unlikely to act n The rate at which industry and society more
as a hedge against inflation, although it does have a generally adopt carbon-saving technology, including
higher sensitivity to climate change factors than the that which already exists
other commodities and, hence, could be viewed as
protection against climate risks. More information on n Cost curves and relative prices of oil, gas and coal
the supply-and-demand side drivers of the carbon
market includes: n Weather patterns and changes in climactic
conditions that can affect short-term energy
n Government/regulatory policy within and across generation and demand side factors
regions in relation to the level to which carbon
emissions should fall over the long term and the n The rate of economic growth in key countries
path taken to the long-term position
Table 16 presents the sensitivities of timberland,
n The balance between carbon pricing and other agricultural land and carbon to the TIP™ factor risks
methods of reducing emission – such as policy for each scenario. The magnitude (low, moderate, high
oriented towards innovation, research and and very high) of the sensitivity of the asset to the
development and encouragement of various TIP™ risk factors is presented at the top of the table,
mitigation strategies with the colour denoting the direction of the impact
(negative, neutral or positive) for each scenario.
n The rate of breakthrough in a wide range of
technologies, including in the cleantech energy/
cleantech sector

Table 16
Sensitivities of timberland, agriculture land and carbon to the TIP™ factor risks for each scenario

Asset Timberland Agriculture land Carbon

Sensitivity High High Very high

Regional This scenario is neutral overall for Agricultural prices are expected This scenario is neutral overall
Divergence timberland assets, although some to rise by around 30% under this for carbon, with the participating
regions are leading in promoting scenario due to climate change,45 regions leading to a rise in the
sustainable forestry and alignment which is not dissimilar to the other price of carbon to as high as $110/
with the REDD and REDD+ scenarios out to 2030. The regional tCO2e (for participating regions
frameworks. Some examples of differences become magnified and industries only). By 2020, there
adaptation finance include the Forest beyond 2050 with an increase will still be different carbon prices
Carbon Management programme in global unrest and geopolitical in different trading schemes and
in Canada, the California Climate risk due to food shortages. There other non-market mechanisms
Action registry in the US, Brazil’s are substantial increases in the that are utilised. In most cases,
Amazon Fund, the Congo Basin risk of hunger among the poorest the carbon emission permits are
Forest Fund and various funds under countries,46 especially in Sub-Saharan allocated free to emitters; by 2030,
the International Climate and Forest Africa and South Asia, where a large there will be linked trading schemes
Initiative supported by countries such portion of the population depends with increasing coverage of global
as Australia and Norway.44 on agriculture, and where capacities emissions.
at the national and farm levels to
adapt to climate change are lowest.

44 For further details, see http://www.forumforthefuture.org/files/Appendix_FIR.pdf.


45 Grantham LSE/Vivid Economics estimates.
46 Parry et al (2004).

66
Table 16
Sensitivities of timberland, agriculture land and carbon to the TIP™ factor risks for each scenario (cont’d)

Asset Timberland Agricultural land Carbon

Sensitivity High High Very high

Delayed Dramatic policy U-turn increases Agricultural prices will rise at a This is neutral for carbon investments
Action the penalties for deforestation similar level to that of Regional until the policy measures are
dramatically, increasing the price Divergence. The dramatic policy introduced, following which time, it
of timberland product prices, turnaround is expected to produce will be very positive with a dramatic
land values and the premium some positive benefits. Policymakers and unanticipated rise in the carbon
attached to carbon trading related promote sustainable agriculture price. By 2020, there will be cap-
activities. Existing timberland assets practices, increasing the reliability and-trade schemes set up in OECD
will appreciate in value and new of long-term crop production but countries, accompanied by taxes and
investments will become more increasing crop production costs in regulation, and by 2030, there will
expensive to invest in. A shift the short term. The delayed action be very high costs of mitigation, with
towards sustainable forestry products also increases risks of climate change global prices as high as $220/tCO2e.
will be demanded by customers. damage, meaning some regions may
Compliance and monitoring costs experience a reduction in available
with policies will increase. land for cultivation (Latam, SE
Asia and Africa)47 – hence, neutral
overall.

Stern Climate policy creates incentives to This is the most positive scenario This is a very positive scenario for
Action reduce deforestation and protect for agricultural investments, as carbon, with the supportive policy
native forests via REDD and REDD+. prices are expected to rise in similar environment increasing the carbon
The demand for sustainably fashion to the other scenarios, price and its relevance to business
harvested forest resources may but the global policy efforts and practices across industries and
increase to fulfil the growing need efficient policy approach promote regions. By 2020, there will be
for timberland. Policies increase the sustainable crop methods, reducing linked OECD trading schemes, and
demand for sustainable forestry the risk of disrupted production. emission trading schemes will also
products.48 Existing assets perform Substantial capital is available to be introduced in BRIC countries
strongly and new investments are assist countries in adapting to (Brazil, Russia, India, China), with
more expensive as land values and climate change in farming methods. the majority of allowances auctioned
timberland costs rise. Sustainable farming and heat- plus complementary taxes and price
tolerant and drought-tolerant crops measures. The carbon price will be
will be introduced, improving climate $110/tCO2e.
resilience and production reliability.

Climate Changes in forest productivity are Climate change physical impact risks EU ETS phase II will end by 2012.
Breakdown likely due to an increased risk of increase, potentially reducing the By 2020, there will be a new EU
degraded ecosystem services that availability of prime cropland but ETS and possibly regional trading in
will require new strategies for increasing the availability of marginal some US states. There will be a low-
management and adaptation. In cropland. The beneficial effects are carbon price ($15/tCO2e) and a small
particular, climate change could likely to be experienced in North coverage of global emissions. By
create water supply concerns in America and Russia, with the biggest 2030, the share of global emissions
regions where tree plantations are losses in Africa and Latam. Risk of in carbon trading falls as non-OECD
most productive. protectionist policies in response to emissions rise.
food shortages could create unrest
and additional geopolitical risk
premium for agricultural investments.
Source: Mercer, drawing from various sources, as referenced

47 Grantham LSE/Vivid Economics, unpublished research adapted from Fischer et al (2002).


48 World Resources Institute Annual Report 2008, available at http://www.wri.org/publication/wri-annual-report-2008 pdf.

67 Climate Change Scenarios – Implications for Strategic Asset Allocation


68
Country and
regional impacts
The following discussion summarises the
analysis that underpinned the conclusions
regarding the investment impacts of
climate change for each country and
region. As for the asset classes, the analysis
is largely qualitative in nature, requiring
judgement and interpretation of the climate
change risks and evidence as presented in
this report.

69 Climate Change Scenarios – Implications for Strategic Asset Allocation


A few key assumptions and considerations: slowing in the pace of investments due to political
impasse. By 2030, additional cumulative technology
n There are limitations of country coverage due to investment inflows could reach $1.3 tr in the best-
data availability, although the countries included case Stern Action scenario, or a lower $650 bn in the
are the largest in terms of technology investment more likely Regional Divergence scenario.
markets and emissions levels.
n “Laggard” – There is little indication that Russia
n The rate of change in technology investment is going to be at the forefront of technology
and climate policy is moving quickly; hence, the investment, where we expect incremental
conclusions drawn in this discussion require cumulative investment flows into technology
periodic review and updating. to be a modest $35 bn by 2030 in the Regional
Divergence scenario.
n Country-level risk on each TIP™ risk factor might
not capture the developments at the sector Impacts
level within countries, such as supportive policy n Impact risks are highest in the Climate Breakdown
measures in buildings, or renewables, for example. scenario, particularly for India/South Asia. This
While such measures might still position a country could destabilise the market and increase the
as less well-positioned for transformation than premium demanded by investors. Total adaptation
others at the aggregate level, these opportunities and residual damage costs in India/South Asia are
will emerge and need to be considered on a case- estimated to be $71 bn, or 0.9% of the level of GDP
by-case basis. by 2030, increasing to $309 bn or 0.6% of the level of
GDP by 2050.
Some highlights of the country/regional analysis for
each factor are summarised below. Policy
n Policy risks are greatest around the Delayed Action
Technology scenario for all regions, as the higher level of
n “Leaders” – The regions that are best positioned to emissions and the higher costs of delayed policy
capture the technological transformation are the EU create instability. This suggests that a delayed
and China/East Asia, as both regions move forward policy response is costly for all countries and
to reduce emissions and attract investment at a regions – there are no winners, as they all face
faster pace than the other countries. In the EU, the the future (higher) adjustment costs, with the
additional cumulative investment levels versus BAU higher cost hitting China particularly hard given
could reach $1 tr by 2030. In China, it could reach its trajectory of rising emissions. We estimate that
$1.3 tr by 2030, making it the largest low-carbon the policy delay may increase adjustment costs
investment market in the world. in China by four times versus the Stern Action
scenario.
n “Improvers” – While the size of investments in
low-carbon energy is comparatively low in Japan The following tables (starting on the next page)
and India/South Asia, it is growing, putting these present the sensitivities of the EU, the US, Japan,
countries in the “improver” category. Incremental China/East Asia, Russia and India/South Asia to the
cumulative investment flows into technology are TIP™ factor risks for each scenario. The magnitude
estimated to reach $220 bn in Japan by 2030 and (low, moderate, high and very high) of the sensitivity
$450 bn in India/South Asia over the same period. of the asset to the TIP™ risk factors is presented at the
top of the table, with the colour denoting the direction
n “Mature but declining” – The US market is of the impact (negative, neutral or positive) for each
currently one of the deepest in low-carbon energy scenario.49
and efficiency; however, indications point to a

49 As before, all references to future T, I or P factors have been discounted by 3% and the technology data refer to the incremental investment flows versus a business-
as-usual baseline. The technology inflows refer to those areas that will benefit from the low-carbon transformation, such as energy efficiency, renewable energy,
biofuels, nuclear and CCS.

70
Table 17
Sensitivities of the EU, US and Japan to the TIP™ factor risks for each scenario

Asset EU US Japan

Sensitivity Moderate High Moderate

Regional There is low policy risk as one of Opportunities in technology lag Policy implementation risks increase
Divergence the “leading” regions. Additional the leading regions as policy efforts investment uncertainty. Additional
cumulative investment levels versus falter due to political impasse, raising cumulative investment flows of over
BAU are around $1 tr by 2030. uncertainty for investors. Additional $100 bn are seen by 2030, with new
cumulative technology investment opportunities as an “improving”
Achieve GHG reduction goals
inflows accumulate to $650 bn by nation on policy implementation.
of -20% of 1990 level by 2020
2030.
(possibly rising to -30%) and -60% Japan has set policy targets, but
to -80% by 2050. Failure to achieve GHG reduction indications are that these may not
goals equal -17% of 2005 or -4% be met.52 Fail to fully meet GHG
Transformation takes place as a result
versus 1990 levels. emission reduction goals of -25% of
of policy measures, including the cap
on the EU ETS, caps for non-EU ETS Delay in passing the climate change 1990 by 2020 and -60% by 2050.
sectors, incentives for renewables, bill and the movement of public Policies include an increase in
targets for improving efficiency via opinion away from climate policy nuclear power, the reintroduction
building standards, refurbishment, increase the policy risk for investors. of subsidies for photovoltaic power,
vehicle manufacturers and Some states within the US have programmes to make transport more
substantial financial resources for progressive policies50 and continue efficient and spending to promote
green energy programmes, including efficiency in buildings. Substantial
to attract capital.51 There are
CCS demonstration. additional domestic measures
national frameworks, with support at
required to meet the targets. Close
the political level required to increase
monitoring of progress required.
investment. Close monitoring of
progress is required.

Delayed There is higher risk for investors High cost implications for the US Higher costs will also be negative for
Action regarding policy uncertainty, with are likely under this scenario, as the Japan, with political impasse globally
investment flows slowing in low- indications are that the US ranks curtailing policy efforts until 2020.
carbon opportunities due to policy relatively poorly in terms of carbon This reduces investment inflows
stalemate internationally. competitiveness.54 We estimate that by over 30% compared to Stern
Incremental cumulative investment the policy delay increases adjustment Action, with incremental cumulative
flows into technology are estimated costs by a factor of 2.5x versus Stern investments estimated to be around
to be around $700 bn by 2030. This Action. $160 bn by 2030.
is 30% lower than Stern Action and Incremental cumulative investment As for the EU, we estimate that the
Regional Divergence levels. flows are estimated to be around adjustment costs will increase by a
Adjustment costs increase with $900 bn by 2030. This is about a factor of 2.5x that of Stern Action.
higher carbon costs, but the EU third lower than Stern Action levels. However, Japan will be quite more
will be more resilient in responding resilient than many other countries,
The high CO2 intensity of the US
than most other regions, as the EU as it ranks in the top three in terms
economy also means the rise in
countries generally rank highly in of carbon competitiveness.56
inflation and interest rates will hit
terms of carbon competitiveness.53 the US harder as CO2-intensive
economies see a significant increase
in inflation from a carbon-price
shock.55

50 While the United States is not a signatory to the Kyoto Protocol, emissions trading has commenced on a small scale with the Regional Greenhouse Gas Initiative. This
involves states in the northeast of the country, and there is also a proposal to trade allowances between a group of Canadian provinces and US states, largely on the
western seaboard, called the Western Climate Initiative.
51 Deutsche Bank Global Climate Change Policy Tracker: An Investor’s Assessment (October 2009).
52 Vivid Economics (2009). The Carbon Productivity Index shows a significant gap between the reduction in carbon emissions in Japan versus the rate of reduction
required to meet their targets.
53 Vivid Economics (2009): Carbon Competitiveness, Figure 1.
54 Source: Vivid Economics (2009): Carbon Competitiveness by Country, Figure 1.
55 Grantham LSE/Vivid Economics: Mapping Evidence Report, Table 30.
56 Vivid Economics (2009): Carbon Competitiveness, Figure 1.

71 Climate Change Scenarios – Implications for Strategic Asset Allocation


Table 17
Sensitivities of the EU, US and Japan to the TIP™ factor risks for each scenario (cont’d)

Asset EU US Japan

Sensitivity Moderate High Moderate

Stern As for Regional Divergence, with The outlook for the US in this Policy implementation risk in Japan
Action even lower policy risk due to globally scenario is much more positive than declines under this scenario in a
coordinated action. the other mitigation scenarios, as the globally coordinated framework.
policy risk for investors is reduced, Policies include a substantial increase
New investments in technology
allowing investment in technology in nuclear power, subsidies for
where the additional cumulative
to flow. photovoltaic power, programmes to
investment levels versus BAU is
estimated to be around $1 tr by Measures include the long-term make transport more efficient and
2030. extension of the renewable energy spending to promote efficiency in
production tax credit, as well as buildings.
The investment opportunities
tax credits for efficient vehicles and Investment in nuclear, hydro, wind
deepen, with the largest emission
efficiency measures in buildings. and other renewables proliferate.
reductions coming through in
renewable energy (wind, solar) The largest emission reductions come The additional cumulative investment
that will continue to dominate through the building, transport and level versus BAU is estimated to be
the market, with markets related biofuel sectors. Renewables and CCS over $220 bn by 2030.
to energy efficiency in buildings, expand considerably. The additional
and transport, nuclear and cumulative investment level versus
commercialisation of CCS. BAU is estimated to be around $1.3
tr by 2030.

Climate The risks of rising impact costs may There are some possible benefits Total adaptation and residual
Breakdown increase within the EU for climate- from climate change for the US, such damage costs are estimated to be
vulnerable regions, such as southern as increasing cereal yields in parts of $10 bn, or 0.1% of the level of GDP
Europe, where extreme heat, fire North America. Coastal areas price in by 2030, increasing to $23 bn, or
and drought risks increase. Total flood risk in major cities and extreme 0.1% of the level of GDP by 2050.59
adaptation and residual damage weather events. Total adaptation and
There is no additional investment in
costs are estimated to be $18 bn, or residual damage costs are estimated
technology related to low-carbon
0.1% of the level of GDP by 2030, to be $64 bn, or 0.4% of the level
beyond BAU.
increasing to $48 bn, or 0.3% of the of GDP by 2030, increasing to $150
level of GDP by 2050.57 bn or 0.7% of the level of GDP by
2050.58
With concerns about industrial
competitiveness on the rise, the EU Federal plans to trade emissions in
ensures that firms covered by its ETS the US founder in Congress, which
face a generous cap on emissions, proves a major blow to global
depressing the carbon price. ambitions on climate change.
There is no additional investment in
technology related to low-carbon
beyond BAU.
Source: Mercer, drawing from various sources, as referenced

57 Grantham LSE/Vivid Economics estimates, using the PAGE2002 model.


58 ibid.
59 ibid.

72
Table 18
Sensitivities of China/East Asia, Russia and India/South Asia to the TIP™ factor risks for each scenario

Asset China/East Asia Russia India/South Asia

Sensitivity High Moderate Moderate

Regional Policy risk is low, as China is also Incremental cumulative investment The policy implementation risk in
Divergence a “leading” region under this flows into technology are estimated India/South Asia increases uncertainty
scenario, with additional cumulative to be a modest $35 bn by 2030. for investors, as progress so far on
investment levels versus BAU rising improving carbon productivity has
Russia announced an intended
to over $1.3 tr by 2030, making it been slower than for other regions.60
reduction in emissions, relative to
the largest low-carbon investment 1990, of 10%–15% by 2020. This The size of investments in low-
market in the world. represents a substantial increase in carbon energy is comparatively low,
China achieves its national climate emissions relative to today’s level and but growing, putting India in the
plan and goal to cut emission puts Russia in the “laggard” higher “improver” category.
intensity by 40% to 45% from 2005 risk category for investors. Incremental cumulative investment
to 2020. In the absence of a framework and flows into technology are estimated
National policies would be policy efforts to reduce emissions, to be around $220 bn by 2030,
implemented, increasing investment along with continued reliance on which is around 2% of India’s
opportunities in nuclear and fossil fuel energy sources, investment projected GDP.
renewables in power generation in technology will remain low. India goes some way to achieving
(including CCS), along with its aim to reduce emission intensity
opportunities related to rebalancing from 2005 to 2020 by 20% to 25%.
the Chinese economy towards
Opportunities are highest in wind,
services and standards for building
due to a government-imposed
efficiency.
renewable portfolio standard, which
starts at 5% in 2010 and increases
to 15% by 2020.

Delayed As the world’s future largest emitter As the 17th largest emitter of energy Higher adjustment costs for India/
Action of CO2 under this scenario, China CO2 per capita in 2007, Russia’s South Asia are also expected, with
would bear the highest adjustment failure to reduce carbon emissions India ranking in the bottom 3 in
costs of all the regions under this will increase adjustment costs terms of carbon competitiveness.62
scenario. considerably under this scenario. We estimate the policy delay will
increase adjustment costs by a factor
We estimate the policy delay will We estimate the policy delay will
of 2.5x versus Stern Action.
increase adjustment costs by 4x increase adjustment costs by a factor
versus Stern Action. of 2.5x versus Stern Action. Incremental cumulative investment
flows into technology are estimated
Incremental cumulative investment Incremental cumulative investment
to be around $350 bn by 2030.
flows into technology are estimated flows into technology are estimated
This is 20% lower than Stern Action
to be over $1 tr by 2030. This is still to be around $140 bn by 2030.
levels.
substantial but around 30% lower This is about 20% lower than Stern
than Stern Action levels. Action levels. India/South Asia may also be riskier
for investors due to higher impact
Some studies also point to a As for the US, the high CO2 intensity
risks associated with physical
potential risk of physical damage to of Russia means the rise in inflation
changes to the environment, with
the environment due to policy delay, and interest rates will hit Russia
increased risk of flooding, drought
including flood risk and disruption to harder as CO2-intensive economies
and disruption to water supply.
water supply.61 see a significant increase in inflation
from a carbon-price shock.

60 Vivid Economics (2009): Carbon Productivity, Figure 3.


61 Yohe et al (2006) identifies China and Argentina among the most vulnerable individual countries. However, other studies place China as a lower risk.
62 Vivid Economics (2009): Carbon Competitiveness by Country, Figure 1.

73 Climate Change Scenarios – Implications for Strategic Asset Allocation


Table 18
Sensitivities of China/East Asia, Russia and India/South Asia to the TIP™ factor risks for each scenario (cont’d)

Asset China/East Asia Russia India/South Asia

Sensitivity High Moderate Moderate

Stern Globally coordinated action further Russia is the only country included Policy slippage risk in India declines
Action reduces policy risk for investors in in this study, which may prove to be under this scenario in a globally
China. a higher risk for investors under this coordinated framework. India also
scenario. benefits from adaptation finance
New investments in technology
from developed markets to help it
where the additional cumulative Russia’s heavy reliance on high-
prepare for future damage due to
investment levels versus BAU is carbon energy-intensive industries
climate change.
estimated to be over $1.3 tr by and lack of preparedness in terms
2030. of reducing emissions will be Investment expands in nuclear power
costly, even in an efficient policy plants and renewables in power
The investment opportunities
framework. generation, particularly hydro, wind
deepen, particularly in wind, solar,
and solar. There will be policies to
hydro, CCS commercialisation Modest investment opportunities in
promote cleaner transport, including
and other renewables and energy technology emerge in the transition,
the use of mass transport and
efficiency measures. The imposition where the additional cumulative
more efficient cars. Continue the
of a feed-in tariff and an abundance investment levels versus BAU is
implementation of CDM projects and
of low-interest state bank loans, estimated to be over $180 tr by
expand CDM to more sectors.
along with cheap turbines, continue 2030. The key areas will be energy
to fuel a surge in wind investment. efficiency and investment in nuclear, The additional cumulative investment
In the solar photovoltaic industry, renewables and CCS. levels versus BAU is estimated to be
manufacturers increase their share of over $450 bn by 2030.
global production considerably.

Climate The absence of investment in low Vulnerability is lower in Russia, for In India/South Asia, risks related to
Breakdown energy infrastructure solutions in whom initial changes in climate are physical damage to the environment
this scenario could thwart its ability likely to be beneficial on aggregate resulting from the lack of policy
to sustain economic growth, with as crop yields increase in response to action will be the highest of the
increased pressure on resources from rising temperatures. countries included in this study,
population growth and rising living particularly water pressures and flood
However, as for China, continued
standards. risk, which could destabilise the
reliance on fossil fuels is associated
China’s reliance on fossil fuels grows market and increase the premium
with an increase of emissions of 1.5x
rapidly – an increase in emissions of demanded by investors.
versus Stern Action levels to 2030.
over 2.5x versus Stern Action levels This increases future carbon costs Emissions of energy-related CO2 in
to 2030. This significantly increases and risks for investors if/when a cost India are 1.5x the level they would
the future carbon liability for China of carbon is enforced further in the be versus Stern Action – hence,
and, hence, risks for investors if/ future. carbon risks increase for future policy
when policymakers do respond measures that enforce a carbon price
Total adaptation and residual
beyond 2030. beyond 2030.
damage costs estimated to be $12
Total adaptation and residual bn, or 0.3% of the level of GDP by Total adaptation and residual
damage costs are estimated to be 2030, increasing to $23 bn, or 0.3% damage costs are estimated to be
$30 bn, or 0.1% of the level of GDP of the level of GDP by 2050.64 $71 bn, or 0.9% of the level of GDP
by 2030, increasing to $76 bn, or by 2030, increasing to $309 bn, or
0.2% of the level of GDP by 2050.63 0.6% of the level of GDP by 2050.65

Source: Mercer, drawing from various sources, as referenced

63 Grantham LSE/Vivid Economics estimates, using the PAGE2002 model.


64 ibid.
65 ibid.

74
Mapping evidence
to the scenarios
This section was produced by Grantham
LSE/Vivid Economics. It formed the basis for
Mercer to later develop and formulate the
TIP™ factor risk framework to translate the
evidence into impacts for asset allocation.

75 Climate Change Scenarios – Implications for Strategic Asset Allocation


Uncertainties around the
outcomes
Some uncertainties for the investor originate in climate
physics. One of the most important sources of risk for
assessing the potential economic impacts is a change
in precipitation patterns, which is also particularly
hard for climate scientists to predict. Furthermore, the
most consequential changes may come from changes
in the frequency and severity of extreme precipitation
events, particularly drought. These extremes are even
less well-understood than the averages both because the
climate simulations have not been run sufficiently to
characterise infrequent events and because it is harder
to calibrate the models for these events.

Another concern for the investor is the widespread


assumption in the literature that policies are optimal,
strategies efficient and institutions robust. In reality, the
delivery of adaptation and mitigation will fall short of
the ideal.

In addition to uncertainties around the scientific


evidence and the likely shape and form of climate
policy measures, investors also face uncertainties
related to the rate of change in technology development
and deployment and what this means for their
investments in different businesses, industries and
regions in the future.

The uncertainties present today, coupled with the


surprising smallness of scale of the research activities
exploring these questions and the degree of difficulty
of the tasks confronting them, suggest that uncertainty
will be an enduring feature of climate change for
investors for some time yet.

76
Macroeconomic impacts

The total impact of climate change on economic neutral for all other scenarios. In Delayed Action,
output can be broken into three contributory factors: it is assumed that a carbon tax (or its equivalent)
is introduced and not fully anticipated; thus, the
n Mitigation costs: the added costs of reducing inflationary effect of a carbon-price shock can be
greenhouse gas emissions considerable, with Grantham LSE/Vivid Economics
estimating it to be in the range of 0.6%–2.1% higher.
n Adaptation costs: the added costs of adapting For the purposes of asset-allocation assumptions,
economies to climate change (for example, by we would recommend some caution interpreting
heightening sea defences) these results, as the inflation impacts would vary
by region; hence, having an inflation increase in
n Residual damage costs: adaptation may not the midpoint of this range under Delayed Action
entirely eliminate the economic costs of physical is reasonable, with inflation remaining unchanged
climate change; hence, this represents the residual across the other scenarios.
damage to the physical environment in addition to
adaptation costs n Interest rates: Using a simple model of central
bank behaviour by applying a coefficient of 1.5 on
Grantham LSE/Vivid Economics applied the World inflation using the Taylor Rule results in a potential
Induced Technical Change Hybrid (WITCH) model to initial increase in central bank interest rates in
estimate the macroeconomic impacts of these three the range of one to three percentage points under
factors for the Stern Action and Climate Breakdown Delayed Action. For the other scenarios, there is no
scenarios, describing it as a “top down” model that impact on interest rates. For the purposes of asset
has considerable technological detail. It is also multi- allocation, we have assumed a rise in the risk-free
regional. For the Regional Divergence and Delayed rate at the lower end of this range for the Delayed
Action scenarios, they applied sensitivity analysis Action scenario, with interest rates remaining
to explore the future potential outcomes from each unchanged for the other scenarios.
scenario.67 In climate-change economics, the impacts
of physical climate change, adaptation and mitigation n Investment uncertainty: The degree to which each
on GDP growth are conventionally expressed as the scenario may create uncertainty for investors varies
percentage difference in the level of GDP, relative to significantly across the scenarios, depending on the
a baseline, in a particular year. Table 19 (on page 78) rate of transformation to a low-carbon economy
summarises the results on that basis. and the timeliness, transparency and level of global
coordination around climate policy. The uncertainty
n GDP impact: The results show that the level of GDP is highest under Delayed Action, where investors
for the Delayed Action scenario would be 5% lower do not fully anticipate the changes, followed by
than it would otherwise have been in 2050, in the Regional Divergence. Stern Action is the scenario
absence of efforts to cut carbon emissions. According that provides the most clarity for investors within
to Grantham LSE/Vivid Economics, this would the horizon of this study, while Climate Breakdown
translate to a decline in annual average growth by presents the greatest long-term risk as the economic
around one-tenth of one percentage point every year impacts of climate change increase significantly
to 2050. From an asset-allocation perspective, this beyond 2050 (see Box 1 on page 79). As highlighted
cost is not significant enough to justify changing earlier in this report, as with systemic risks in the
the asset-class assumptions related to GDP growth past (the IT bubble, credit crisis), the source of
across the climate scenarios in that period. uncertainty for investors over the next 20 years
is likely to come from unanticipated events and
n Inflation impact: The long-run equilibrium results the way the market behaves in response to such
showed a potential inflationary impact under developments, rather than being led by changes to
the Delayed Action scenario, with inflation being long-run macroeconomic outcomes.

67 This was delivered to Mercer and the project group as the “scenarios report” for this project.

77 Climate Change Scenarios – Implications for Strategic Asset Allocation


Table 19
Uncertainty and macroeconomic impact

Scenarios Degree of investment GDP impact Inflation impact Interest rates


uncertainty (% change from GDP (% change CPI) (% change cash rate)
level)

2030 2050 2030 2050 2030 2050

Regional Impact varied by -1.2 -3.9 Neutral Neutral Neutral Neutral


Divergence regions, with leaders
and laggards creating
higher uncertainty

Delayed High level of uncertainty -1.3 -5.2 +0.6% to 2.1%, +0.9% to 3.2%,
Action before policy changes, varies by region varies by region
which are not
anticipated; uncertainty
declines following policy
measures

Stern Low uncertainty due -1.1 -4.3 Neutral Neutral Neutral Neutral
Action to climate policy
transparency that
is coordinated and
anticipated

Climate Low uncertainty -0.5 -1.0 Neutral Neutral Neutral Neutral


Breakdown until 2050, but then
increasing, possibly
abruptly
Source: Grantham LSE/Vivid Economics estimates based on mitigation, adaptation and residual damage costs

The GDP estimates are in line with those made in the


Stern Review, which used the more standard method of
expressing costs in terms of the level of GDP. The large
estimates produced by the Stern Review of the physical
impact of climate change are driven, in large part,
by what happens after 2050 and, indeed, after 2100.
Box 1 (on page 79) explains this. However, due to the
inertia in the climate system, we need to cut carbon
emissions in the near term in order to avoid these
impacts in the long term. One should also bear in mind
that the models used to estimate the costs of physical
climate change in particular are widely understood
to be imperfect, and some have suggested that they
underestimate the economic cost of climate change.

78
Box 1:
The full and long-run economic cost of climate change

The residual damages of climate change are at the level of individual countries. Small economies
calibrated on so-called “market” sectors such as in particular have proved in the recent past to be
agriculture, energy and forestry. The distinguishing vulnerable to extreme weather events. The Stern
feature of these sectors is that goods and services Review (Stern, 2007) gives several examples, including
have market prices, and so climate damage has a real the 1991–1992 drought in Zimbabwe, which led to
effect on economic output and, therefore, potentially a doubling in the country’s current account deficit
on other macroeconomic variables, which is the focus and in its external debt. Another is Hurricane Mitch,
of this study. which caused devastation in Central America in 1998
– Honduras, for example, faced reconstruction costs in
However, the full cost of climate change on economic excess of national GDP.
welfare extends beyond these sectors to take in
impacts on so-called “non-market” sectors such as But perhaps the most important issue that is not
natural ecosystems and human health (over and reflected in the analysis is the impact of climate
above effects on labour productivity). These impacts change in the longer run. Studies from the level of
are valuable to human beings, to the extent that particular sectors, such as agriculture and health
they are willing to pay to avoid them, or willing to through to global economic costs, virtually all agree
accept compensation for them (that is, the concept that impacts would become predominantly negative
of the value of a statistical life). But they are not and increase rapidly in magnitude after 2050. Yet since
experienced as measurable changes in macroeconomic many of the greenhouse gases emitted today will
performance, because market prices do not exist. still reside in the atmosphere until 2100 and beyond
This is one reason why the estimates necessarily (particularly CO2), emission reductions are required
understate the true welfare cost of climate change. in the short term in order to avoid them. Figure 10
Using central estimates, PAGE2002 projects that over makes the point about long-run impacts, presenting
half of the welfare cost of climate change for 2.5°C the market impacts of climate change for the period
warming is due to these “non-market” damages. 2050–2200, as estimated by the PAGE2002 model in
the Stern Review (Stern, 2007). Notice that 2050 is the
Another factor that is ignored in the above, due to origin in this chart. The cost of climate change rises
the effect of averaging across large world regions, is rapidly after 2050.
the possibility of strong and direct economic impacts

Figure 10

2050 2100 2150 2200


0

-5
-7.3
-10
% loss in GDP per capita

-15

-20

-25

-30 High climate, market impacts + risk of catastrophe


5%–95% impacts range
-35

-40

Source: Stern 2007, Grantham LSE/Vivid Economics

79 Climate Change Scenarios – Implications for Strategic Asset Allocation


Commodity prices not significantly so until mid-century. That is due to
the fact that physical impacts under the two scenarios
Commodities could be affected by a number of do not start to diverge appreciably until mid-century.
factors, including changes in supply and demand due
to mitigation policies, and changes in supply due to Carbon price
physical impacts. The climate models examined for this study show that
the key factors determining the carbon price are:
Fossil fuels
Two things can be expected to happen to fossil-fuel n The ambition of attempts to mitigate climate change
prices under an ambitious mitigation scenario. First, (the ultimate atmospheric concentration for which
demand for fossil fuels would be lower than in the policymakers aim)
climate breakdown scenario, depressing prices received
by fossil-fuel owners (with the carbon price acting as n The flexibility with which emissions reductions can
a wedge between the prices received by producers and be made, the extent and timing of coordinated global
the prices charged to customers). Second, the cost to mitigation policies, and the inclusion of as many
customers of the most carbon-intensive fuels would sectors as possible under a single carbon pricing
increase the most following the imposition of a carbon regime
price, causing a relatively greater drop in demand
and lowering prices to the owners of those fuels to a n The availability of different technological options,
greater extent. which may be constrained by either political or
physical feasibility; rates of technological innovation
In the case of Regional Divergence, the geopolitical and the possibility of breakthrough technologies;
situation is most likely one where individual countries the degree of foresight economic actors have about
or regions don’t want to be dependent on the supply future abatement options and costs
from other regions. This would trigger a drive to self-
sufficiency that should decrease the fossil fuel demand n Fossil-fuel prices and an offsetting effect, whereby
by the western world. reduced demand for fossil fuels (for example, due to
recession) lowers the cost of emitting and dampens
Agricultural commodities carbon prices
Crop yields are expected to be higher in many
temperate regions, but lower in most tropical regions. n The ease with which energy inputs to production
We would therefore expect, all else being equal, to see can be substituted
corresponding price increases (decreases) for crops
grown in the relevant regions as supply is reduced n The existence of major policies other than carbon
(raised), with these two effects on commodity price pricing – for example, large-scale renewables quotas,
indices counteracting each other to some extent. The as in the EU, could reduce demand for carbon credits
regions that would see a crop yield increase (North by forcing certain mitigation actions
America, Russia) are also important for global food
production. However, population is also expected to Flexibility in emission reductions is vital for
rise, increasing the demand for food and exerting maintaining low costs, through taking advantage of
an upward pressure on prices under both scenarios. cheaper emission reductions wherever and whenever
That could be offset to some extent by improved crop they can be made.68 Policy regimes that do not achieve
varieties and agricultural practices increasing yields. this flexibility because they segment the carbon
The price impact attempts to separate the effects regime, either regionally or sectorally, see higher
of climate change from other factors driving price carbon prices.
changes, based on forecasts from the studies of Fischer
et al (2002) and Parry et al (2004). Commodity prices are Unsurprisingly, therefore, the carbon prices that
expected to be higher under climate breakdown, but emerge from analyses are highly sensitive to any

68 And reductions of the greenhouse gases that are cheapest to abate at the margin. “Carbon pricing” is a convenient shorthand for pricing greenhouse gas emissions
in general, with each GHG emission price having an “exchange rate” with the literal carbon price.

80
barriers to flexibility that are imposed. Estimated estimates increase more steadily with time. Prices
carbon prices at different points in time vary hugely accelerate because the model’s agents have perfect
between models but are typically in the range of a few foresight; they require a relatively modest carbon
tens to a few hundreds of dollars by 2030 for scenarios price to take early action (given their expectation of
with the ambition of Stern Action. higher rises subsequently), but they expect limited
technological options and substitution possibilities
The price estimates provided in Table 20 for 2030 are within the energy sector later, because cheaper options
derived by Grantham LSE/Vivid Economics from the are exhausted earlier on. Nearly all projections of
WITCH model of the RECIPE study and are based on a carbon prices entail a period-by-period increase for
global carbon trading regime. They are in the middle several years, often well into the second half of the
of model estimates in the RECIPE study over the time century or beyond.
period, but increase rapidly after 2030, while the other

Table 20
Commodity price impact to 2030

Scenarios Carbon price Oil price Coal Gas Agriculture


($/tCO2) (% price change) (% price change) (% price change) (% price change)

2030 2030 2030 2030 2030

Regional 110 N/A N/A N/A N/A


Divergence

Delayed 220 N/A N/A N/A N/A


Action

Stern 110 -2.0 -36.6 10.4 +37


Action

Climate 15 25.2 60.0 35.4 +38


Breakdown

Source: Grantham LSE/Vivid Economics estimates


Agriculture refers to percentage difference versus 1990 levels due to climate change rather than other effects on agricultural prices.
Oil, coal and gas prices refer to 2008 dollars per GJ energy provided.

81 Climate Change Scenarios – Implications for Strategic Asset Allocation


Technology investment

Energy supply and the fuels and technology mix observed in CCS (which has no economic justification
that deliver it would be driven by many interacting under a “no mitigation” scenario) and non-biomass
factors, key among which are economic growth, new renewables (wind, solar and hydro-electric power).
innovations related to research and development
expenditure, population growth, fossil-fuel prices Energy efficiency and decarbonisation
and any policies put in place to reduce both energy
demand and the carbon intensity of energy supply. All Energy efficiency could have as much, or even more,
of the models examined in this study show profound potential to abate emissions than switches to low-
switches away from fossil-fuel production in ambitious carbon technologies. Figure 11 illustrates the potential
mitigation scenarios. abatement in 2030 attributable to energy efficiency
and decarbonisation of the energy supply, according
In 2030, about two-thirds of the shift in fossil-fuel use to the RECIPE WITCH model that was built by a team
is attributable to lower overall energy demand, while from the FEEM research institution, the International
the remaining third results from supply-side changes. Energy Agency’s (IEA) World Energy Outlook 450
Supply-side differences result from the growth of ppm mitigation scenario and the McKinsey global
renewables, biomass and nuclear power, as well as the abatement cost curve. In all three, abatement from
entry of CCS as a viable technology, under Stern Action. energy efficiency is as great, or slightly greater, than
The greatest differences across the scenarios are from decarbonisation of energy supply.

Figure 11
Energy efficiency accounts for at least half of potential abatement

Decarbonisation
Energy efficiency

30
Abatement in 2030 under ambitious mitigation

25

20
scenario (Gt CO2e)

15

10

0
RECIPE WITCH IEA WEO McKinsey
Source: Grantham Research Institute LSE and Vivid Economics, based on Edenhofer (2009), IEA (2009) and McKinsey and Company (2009)

82
Technology deployment deployment. To understand the set of measures likely
to be necessary, Figure 12 depicts the innovation
It is widely understood that a carbon price would process. Governments have a role to play in the early
be insufficient in itself to bring about the large- stage R&D of low-carbon technologies, in particular
scale development and deployment of low-carbon through public subsidies. As technologies move closer
technologies, because of the existence of other to the market (especially the diffusion stage), private
barriers in this area (for example, see Stern 2007). returns increase, as do the required flows of finance.
These barriers include the spillovers to innovation of Here, private-sector investment takes the lead, but
new technologies, which the innovating firms cannot governments still have a role to play in providing a
capture, as well as problems of “lock in” of existing credible, long-term policy framework. In between,
technologies, due in large part to returns to scale. during demonstration and deployment, the picture is
These barriers are particularly acute in key sectors for more complex and a variety of regulatory measures
climate change, such as energy supply and transport. might be applied – as will be discussed further in the
next section.
Therefore, public support and regulatory measures
are required to support technology development and

Figure 12
Governments push innovation of new technologies, markets pull them

Governments Markets

Financing
technology,
and ideas
Number of projects

Research
and Demonstration Deployment Diffusion
Development

Source: Grantham Research Institute/Vivid Economic, base Feedback


on World Bank (2009b)

83 Climate Change Scenarios – Implications for Strategic Asset Allocation


Sector impacts

Mitigation Sectors that are highly sensitive to mitigation policy


include agriculture, forestry and fisheries, chemicals,
The costs of mitigation may not be evenly distributed primary metals, crude oil and gas extraction, and
across sectors of the economy. In general, those sectors metal mining. Their vulnerability derives from the fact
with fossil fuels for output or with the highest carbon that these sectors face a direct cost due to their own
intensity of production, and those whose output can greenhouse gas emissions, and an indirect cost due to
be most readily substituted for, would suffer the most their high energy consumption.
from the introduction of carbon pricing. Sectors with
high-carbon intensities would face higher costs, while Sectors with medium sensitivity to mitigation include
sectors whose goods can easily be substituted for tend the bulk of manufacturing: motor vehicles, glass and
to have a demand that is very responsive to increases minerals, pulp and paper, machinery manufacturing
in price. and so on. Some of these sectors are relatively energy-
intensive, but there may be relatively few or imperfect
Using Goettle and Fawcett (2009) to make a qualitative substitutes for their products.
ranking of the sectoral impacts of mitigation policy
(Table 21), we find that sectors with very high Sectors with low sensitivity to mitigation policy
sensitivity are, unsurprisingly, those with fossil fuels as are those with low energy use, low consumption of
their output – that is, coal mining, petroleum refining carbon-intensive inputs, and little in the way of direct
and gas utilities. These sectors face a significant combustion. Services dominate this category, but it also
decline in output, because a carbon constraint includes some less energy-intensive manufacturing.
makes the goods that these industries provide more Indeed, any declines in output among these sectors
expensive than low- or zero-carbon alternatives, while are generally not a result of an increase in the cost of
consumers can easily substitute a joule of energy from production, but rather the effect of a decline in real
renewable sources for a joule of energy from fossil-fuel incomes, which leads consumers to consume less of all
sources. Electricity utilities are also highly sensitive, as goods in the economy.
the overall cost of supplying electricity increases and
demand correspondingly falls.

Table 21
Energy- and carbon-intensive primary and manufacturing industries would suffer the biggest impact of a carbon constraint

Sensitivity Sectors
category

Very high Coal mining; petroleum refining; gas utilities (services); electric utilities (services)

High Agriculture, forestry and fisheries; chemicals and allied products; primary metals; crude oil and gas extraction; metal
mining; non-metallic mineral mining

Medium Stone; clay and glass products; fabricated metal products; non-electrical machinery; paper and allied products;
electrical machinery; motor vehicles; rubber and plastic products; furniture and fixtures

Low Lumber and wood products; transportation and warehousing; wholesale and retail trade; construction; other
transportation equipment; leather and leather products; instruments; miscellaneous manufacturing; printing and
publishing; government enterprises; apparel and other textile products; finance; insurance and real estate; personal
and business services; communications; textile mill products; food and kindred products; tobacco manufacturers
Source: Grantham Research Institute/Vivid Economics, based on Goettle and Fawcett (2009)

84
There would, of course, also be winners from output most depends on prevailing weather conditions,
mitigation policy. Analyses of industrial sectors are such as agriculture, forestry and water. Coastal-zone
generally not able to detect these winners, because economic activity is vulnerable to sea-level rises,
they tend to be types of business that fall within, especially floods and storms.
or cut across, traditional sectoral classifications.
While different studies make different assumptions As with mitigation, it is also to be expected that the
about technological possibilities, it seems clear that impacts of climate change would trigger second-round
winners would include renewable and nuclear power effects on the performance of other sectors of the
supply firms, the part of the agriculture sector that economy that demand goods and services from sectors
specialises in biofuels, and firms supplying CCS and directly affected. However, there has been little or no
energy efficiency technologies (for example, smart- research on these second-round effects, and we are
grid components and energy-use auditing methods). limited to considering the first-round impacts only.
These are discussed further in the Listed Equities and
Renewable Energy investment impacts in this report. Table 22 summarises the present state of knowledge
about positive and negative impacts of up to 3°C global
Impacts of physical climate change impacts warming on different sectors. The evidence is drawn
from the synthesis report of the Intergovernmental
There may also be winners and losers from the Panel on Climate Change (Parry et al, 2007). The pattern
impacts of climate change itself. To a first order, the is mixed and region-specific, except for coastal zones,
sectors that stand to gain and lose the most from where the costs of protection and residual damage are
physical climate change are those sectors whose always negative.

Table 22
Initial climate change has both positive and negative consequences for economic sectors

Sector Sectoral impacts of climate change for up to 3°C warming

Positive impacts Negative impacts

Agriculture Increasing crop productivity in the mid to high latitudes, Decreasing crop productivity in the low latitudes, e.g.
e.g. northern North America, northern Europe, Russia Africa

Forestry Increasing global timberland production overall Decreasing production in northern North America
Increasing production potential in South America

Water Increasing water availability at high latitudes and in the Decreasing water availability in the mid latitudes and in
moist tropics semi-arid, low-latitude regions

Health Decreased morbidity and mortality from cold stress, Increased impact from malnutrition, heat stress, extreme
primarily at mid to high latitudes events, diarrhoea, and some other vector- and water-
borne diseases; burden concentrated on low-latitude
developing regions

Coastal Increased adaptation costs in all regions to protect


zones against flood risk

Energy Decreased requirement for space heating at mid to high Increased requirement for space cooling
latitudes

Tourism Increased tourism in mid- to high-latitude regions Decreased tourism at low latitudes

Source: Grantham Research Institute/Vivid Economics

85 Climate Change Scenarios – Implications for Strategic Asset Allocation


Climate policy

The climate change modelling literature assumes relates to requirements to comply with performance
that reductions in greenhouse gas emissions are standards.
made efficiently. The cheapest emissions abatement
options are always chosen in these models, so that Carbon price instruments
the marginal costs of abatement in different sectors
and regions are equal, as they are across time.69 A foundation of the Stern Action scenario is a
Modelling studies also make assumptions about the coordinated global incentive to abate emissions (for
feasibility of particular emissions reduction techniques example, see Stern 2007) either in the form of an
(for example, efficiency gains) and technologies (for internationally harmonised system of carbon taxes
example, renewable energy and CCS). or in the form of a global market for tradeable
carbon allowances.
The reality investors face is that policy is very unlikely
to be implemented efficiently the way the studies Figure 13 illustrates the state of the global carbon
assume. That was one of the primary aims behind market today. The most well-known ETS is in the EU.
considering some of the wider policy factor risks Established in 2005, it now covers around half of all EU
associated with each climate scenario, as the risk of CO2 emissions and involves trading between firms. In
slippage or inefficient or unanticipated policy action addition, governments signed up to the Kyoto Protocol
could create new opportunities and pose risks for long- and facing binding emissions targets can also trade
term institutional investors. emissions between themselves. Kyoto signatories
outside the EU include Australia, Canada and Japan.70
Three types of regulatory interventions that may Developing countries can sell carbon credits, generated
feature in policy design have been examined in this by emissions reduction projects such as renewable
study. The first involves the introduction of a price energy, to governments with Kyoto targets or to firms
on emissions of carbon, the second is focused on regulated by the EU ETS through the CDM, while
promoting technology development, and the third transition countries can do the same under JI.

Figure 13
The carbon market already has a global reach

Kyoto framework

EU ETS (domestic)

Kyoto signatories outside


EU ETS
Annex I countries with
economies in transition.
Potential JI countries.
Non-Annex I
countries. Potential
CDM host countries.

Non-Kyoto initiatives

Regional Greenhouse Gas Initiative (RGGI) States in the


northeastern United States have also passed carbon
regulations for stationary sources
Western Regional Climate Initiative
California is a leading participant in a regional initiative to
reduce its emissions, along with several Canadian provinces

Source: Grantham Research Institute/Vivid Economics, based on Goettle and Fawcett (2009)

69 There are occasional exceptions to this rule, such as when studies explicitly investigate delayed participation by the developing world in global efforts to reduce
emissions.
70 EU member states can trade emissions both at the government level and at the firm level through the EU ETS to meet their Kyoto targets.

86
In the US, emission trading has commenced on a small substantial emissions reductions. This could either
scale with the Regional Greenhouse Gas Initiative come about through emissions trading or through
(RGGI), involving states in the northeast of the country, harmonised carbon taxes.
and there is also a proposal to trade allowances
between a group of Canadian provinces and US states, Other policy instruments
largely on the western seaboard, called the Western
Climate Initiative. Indeed, there are further proposals A wide variety of measures have been taken around
in several countries that are not included in Figure 13, the world to support the initial deployment of new,
including Mexico, Japan, South Korea, Australia, New low-carbon technologies. Box 2 gives some examples,
Zealand, in Midwestern states of the US, and at the ranging from direct fiscal instruments, such as tax
federal level in the US and Canada. breaks and subsidies, through to quantity-based
schemes like renewables certificates. As recent
For emissions trading to take place at the scale research from Deutsche Bank Climate Change Advisors
necessary to realise Stern Action, current schemes (2009) has shown, deployment support measures for
would have to be expanded, deepened and better low-carbon technologies are proliferating across the
integrated, so that by 2030 at the very latest, there world, in many different forms. This would continue
would effectively be a single global carbon market. under the Stern Action scenario, although measures
The signal of a transition compatible with Stern might be progressively withdrawn as technologies
Action is a rapid geographical spread of carbon mature and are deployed at ever-increasing scales.
prices, accompanied by designs capable of delivering

Box 2:
Examples of deployment support measures for low-carbon technologies

Capital subsidies for demonstration projects and Quota-based schemes, such as the Renewable
programmes, such as for rooftop solar photovoltaic Portfolio Standards in use in many US states, which
panels in the US, Germany and Japan. require electric utilities to source a specified proportion
of their electricity from renewable sources. Such
Tax credits and exemptions, such as the Production schemes sometimes allow the quotas to be traded, as
Tax Credit in the US, which is given to renewable is the case with the Renewables Obligation Certificate
electricity generators in their first 10 years of (ROC) in the UK.
operation, and tax breaks on biofuels in the UK and US.
Tendering for tranches of output, which has been
Feed-in tariffs, which are price premiums paid to used in, for example, China and Canada to ensure
electricity generators to feed renewable energy into renewables make up a certain portion of energy supply.
the grid, have enjoyed notable success in, for
example, Germany and Spain. To be successful, Government procurement policies, ranging from
they are usually combined with a regulatory demonstrator projects on local-government buildings
requirement that eligible renewable electricity to the use of fuel cells and solar technologies in
generators are connected to the grid. national defence and aerospace industries.

Source: Grantham Research Institute LSE/Vivid Economics, based on Deutsche Bank Climate Change Advisors (2009) and Stern (2007)

87 Climate Change Scenarios – Implications for Strategic Asset Allocation


Overcoming barriers to increasing energy consuming asset (for example, a building); and a
efficiency low weight placed on the value of future energy
cost savings (evidence in consumers’ purchasing of
domestic appliances and personal transport). The
Another pillar of climate policy necessary to deliver
introduction of energy performance standards has
the most optimistic mitigation scenario in this
already begun in a few areas and might be used to
study – Stern Action – is a set of regulatory measures
complement carbon pricing in many product areas,
to promote energy efficiency. There are particular
including vehicles (air, road and sea), lighting, boilers,
barriers to increasing energy efficiency that are not
motors and drives, buildings and electrical appliances.
overcome by carbon pricing or technology support
alone. McKinsey and Company (2009) suggests the
existence of abatement options in energy efficiency, Public attitudes, politics and business
which saves money.71 Without the presence of barriers,
these options should already have been taken up, Public support for climate-change mitigation in all
irrespective of climate change. major emitting nations is a precondition for the
Stern Action scenario, given the depth of emissions
These barriers include hidden and transaction costs, reductions required. While a portion of the required
such as the cost of the time needed to plan new emissions reductions may provide co-benefits to, for
investments, the cost of information about available example, energy security and local air quality, and may
options, capital constraints, misaligned incentives thus garner support irrespective of climate-change
such as the landlord-tenant problem,72 as well as goals, the majority of the required reductions would
behavioural and organisational factors. need support in their own right.

While a variety of measures exist to counter such Mounting scientific evidence (for example, on melting
barriers, including information-provision techniques ice caps) is one potential basis for a rise in public
such as energy performance labelling and smart energy concern, but there has been a recent upturn in
metering, perhaps the most important category of climate scepticism in countries such as the UK (BBC,
measure to promote energy efficiency is the traditional 2010), which has happened in spite of the continuing
regulatory standard – for example, mandating the use consensus message from climate scientists.
of particular technologies, banning the use of others, or
mandating minimum efficiency performance. It is possible that local weather events, which are not
conclusively linked to climate change, are a source of
Such regulatory standards relevant to climate-change concern, as they have been in the recent past (Jordan
mitigation are widely used across the world, including and Lorenzoni, 2007). In all nations, rising affluence
in transport (for example, fuel efficiency standards for might lead to an increase in environmental concerns
road vehicles). Expect to see the further proliferation of more generally (for example, see Kristroem and Riera,
such standards under Stern Action, and an increase in 1996), while periods of economic crisis or stagnation
their stringency. could divert attention from the environment.

In contrast to Stern Action, the Climate Breakdown


Standards and process controls scenario is founded on a failure to win widespread
public support. In the immediate future, this is most
Carbon prices and technology policies are unlikely likely to be due to a preoccupation with jobs and
to be sufficient to deliver the Stern Action disposable income, in the wake of the global economic
emissions scenario in areas where the market is downturn, as well as the way in which scientific
unresponsive to price signals. The reasons behind uncertainty plays out in the media and public realm.
price unresponsiveness can be complex. They include High costs of mitigation, as experienced with some
a small cost share for energy, among other costs; a renewables and CCS demonstrations, might deter some
split between the owner and operator of an energy- members of the public.

71 On the other hand, analyses such as Joskow and Marron (1992) contest the idea of negative-cost abatement.
72 The landlord faces the capital cost of an investment in energy efficiency, but the benefit is reaped by the tenant in the form of lower fuel bills.

88
Physical impacts

Taking the two most extreme scenarios in terms to boost sea defences. The sea-level rise from 1990 to
of climate change impacts, the Climate Breakdown 2050 under Climate Breakdown is 0.38 m; under Stern
scenario involves 3°C warming above the pre-industrial Action, it is 0.24 m, with little difference between the
level by 2050, with the Stern Action scenario involving scenarios until after 2030. Table 23 (on page 90) shows
warming f 1.8°C. While this difference in temperature Grantham LSE/Vivid Economics headline estimates
is significant, many of the major differences in climate of the global mean sea-level rise for the two most
impacts between the scenarios arise after 2020, and extreme scenarios in terms of climate impacts –
most of them after 2030. Climate Breakdown and Stern Action. For this study,
the Rahmstorf (2007) projections are used, because
Sea-level rise they appear to provide a better fit of the observational
record of sea-level rises over the past few decades.
One of the main consequences of rising global These projections are also used by the World Bank
temperatures is a rise in sea levels, which increases the Economics of Adaptation to Climate Change study
risk of coastal flooding unless adaptation is undertaken (World Bank 2009a).73

Figure 14
Regional temperature change in 2020 (top) and 2050 (bottom) for BAU

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5


(°C)
Source: Meehl G. et al. “Global Climate Projections” in Solomon S. et al. (eds.) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to
the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. (Cambridge and New York: Cambridge University Press, 2007)

73 Unfortunately, the Rahmstorf projections are only available for the IPCC’s business-as-usual emissions scenarios. In order to estimate sea-level rises under Stern Action,
the data are used to estimate a linear relationship between sea-level rises and global mean warming that can be used to extrapolate to Stern Action. This is likely to be
a reasonable approximation, since Rahmstorf himself fits a linear relationship.

89 Climate Change Scenarios – Implications for Strategic Asset Allocation


Table 23
Little difference in sea-level rises until after 2030

Sea-level rise (metres above 1990)

2012 2020 2030 2050

Climate breakdown 0.08 0.12 0.19 0.38

Stern Action 0.08 0.12 0.17 0.24


Source: Grantham Research Institute/Vivid Economics

Water availability Extreme events

Changes in the availability of water have been Another important way in which climate change
identified as perhaps the most important type of can affect human and economic systems is through
physical change resulting from greenhouse gas changes in the frequency and severity of extreme
emissions (Stern, 2007). Unfortunately, they are also weather. Unfortunately, like changes in water
among the more poorly understood outcomes of availability, this aspect of climate change is generally
climate change. At the regional level, climate models poorly understood. Extreme weather events related
continue to disagree about the direction of changes in to temperature – that is, heat waves and cold snaps –
precipitation over many regions, with some predicting are better understood than storms. The IPCC expects
increases and some predicting decreases. the frequency, severity and length of heat waves to
increase globally (Solomon et al, 2007), with very few
At the same time, most models predict a decrease exceptions. At the same time, a reduction is expected
in precipitation at subtropical latitudes (20–40 in the number of cold snaps affecting the Northern
degrees), especially at the poleward margin of this Hemisphere. This, in turn, is expected to reduce
belt. Precipitation is expected to fall over Central morbidity and mortality due to cold and increase the
America, with the southern states of the US falling length of the growing season for agriculture.
in a more uncertain transition zone. Precipitation is
also expected to fall in southern Europe and North Changes in the frequency and intensity of typhoons
Africa. This drying would be particularly strong in the and hurricanes are very uncertain and forecasts have
summer months, due to greater relative reductions in been the source of controversy. The IPCC tentatively
precipitation and to increased evaporation in the heat. expects an increase in the intensity of typhoons and
hurricanes but a reduction in their frequency (Solomon
In the Southern Hemisphere, very little land in et al, 2007).
the subpolar belt is set to experience increased
precipitation. Instead, areas such as the southernmost Crop yields and arable land supply
countries of Africa and southern Australia may see
decreases in precipitation. Changes in precipitation Agriculture is one of the most sensitive economic
in the tropical belt are the least well-understood, due sectors to climate change. Crop yields are critically
to the complex climate processes in this zone. Most dependent on prevailing temperatures and water
models predict precipitation increases in the summer availability, and can increase as a function of the
monsoon season in South and Southeast Asia, as well atmospheric concentration of CO2 (CO2 is an input to
as in East Africa. plant growth).74 Based on Parry et al (2004), the chart
below maps the global change in crop yield due to
climate change in 2020 and 2050 on a BAU scenario
similar to the Climate Breakdown scenario.75 Crop
yields are defined as the aggregate yield of wheat, rice,
maize and soybean.

74 Climate change is also expected to affect crop yields indirectly through, for example, plant and insect pests.
75 Parry et al use the AOGCM of the UK’s Hadley Centre. When run with the IPCC’s high emissions A1FI scenario, the Hadley Centre model forecasts an increase in the
global mean temperature of approximately 3°C by 2050. Figure 15 includes the fertilisation effect of increased atmospheric CO2.

90
The key message for investors from this chart is Fischer et al (2002) provide estimates of the effect
the significant variation across regions in terms of of climate change on the availability of land with
crop yield impacts. According to Parry et al’s (2004) potential for cultivating major fruit and fibre crops.
projections, under a Climate Breakdown scenario, These are available for major world regions on a
yields would increase in North America, northwestern business-as-usual scenario compatible with climate
Europe, Australia, Argentina, and in parts of Asia. breakdown and are reported for the year 2080.
Conversely, they would decrease throughout Africa and In brief, climate change is forecast to slightly reduce
the Middle East, in Russia and in most of Central and the availability of good and prime cropland in 2050, but
South America. to increase the availability of marginal cropland. The
beneficial effects are experienced in North America
Caution should be exercised in interpreting yield and Russia, with the biggest losses experienced in
projections in particular regions, due to the greater Africa and Latin America. These estimates are in line
uncertainties at this spatial resolution. Nevertheless, with the data for crop yields presented earlier, but
a general conclusion is that climate change under the once again caution should be exercised in interpreting
“no mitigation” scenario causes yields to increase in regional results, given the differences between studies.
most of the currently industrialised world in the period
to 2050, and decrease in most of the developing world.

Figure 15
Percentage change in crop yield in 2020 and 2050 under climate breakdown

-30 -10 -5 -2.5 0 2.5 5 10 20

Source: Parry M., Rosenzweig C. et al. (2004). “Effects of climate change on global food production under SRES emissions and socio-economic scenarios”. Global
Environmental Change 14: 53-67.
Top indicates 2020, bottom 2050.

91 Climate Change Scenarios – Implications for Strategic Asset Allocation


Health impacts and population migration

For institutional investors, health impacts and


population migration can potentially have an impact
on long-term liabilities and affect assumptions around
mortality rates. At present, the evidence available
to consider changes to such assumptions is not
sufficiently strong (as discussed later). The health
effects would be both positive and negative; the
period in which they would become pronounced is
also uncertain. The research on population migration
impacts is sporadic and qualitative, with further
research required to evaluate the potential impact
on pension fund liabilities.76 Grantham LSE/Vivid
Economics highlight that the studies omit potentially
important sources of mortality, including malnutrition
and deaths from extreme events. So they are likely
to be an underestimate of the increase in illness and
death in the period to 2050.

Few studies have quantitatively analysed the effect


of future climate change on global mortality across
multiple diseases. The most comprehensive and
up-to-date study that could be used for this purpose is
Bosello et al (2006). The data show that climate change
would have both health benefits and health costs, and
that these would vary by region. Perhaps surprisingly,
the largest absolute change is a decrease in all regions,
except the EU, in mortality due to cardiovascular
diseases from heat and cold stress. That is, a decrease
in the number of winter cold periods due to climate
change has significant health benefits. Largely on
the strength of this benefit, total global mortality
as a result of climate change falls. On a region-by-
region basis, significant benefits are forecast for China
and India, due entirely to reduced mortality from
cardiovascular diseases (cold stress), as well as eastern
Europe and the former Soviet Union, as well as the US.
Mortality is forecast to increase in Europe and Japan,
but by far the greatest absolute increase is forecast for
the “rest of the world” category, which comprises much
of the developing world. This is due, in large part, to
diarrhoea, but respiratory diseases caused by heat
waves play a role, as does malaria.

76 Brown O. (2008).

92
Methodology
The focus of this study is on the period
to 2030, although the implication of the
scenarios out to 2050 was also considered.

93 Climate Change Scenarios – Implications for Strategic Asset Allocation


The scope of the study in terms of asset classes and
geographies is summarised below.

n Asset classes:

– Listed equities (global, emerging, sustainability,


efficiency/renewables)
– Bonds (government, emerging debt, investment-
grade credit)
– Real estate (core unlisted)
– Infrastructure (core unlisted, efficiency/renewables)
– Private equity (LBO, venture capital, efficiency/
renewables)
– Commodities (agriculture, timberland and carbon)

n Geographies:

– EU
– US
– Japan
– China/East Asia
– Russia
– India/South Asia

To answer these questions a multi-layered, collaborative


approach was taken with the following features:

n Collaboration

n Scenario analysis

n Factor risk framework

n Climate-change factors – TIP™ framework

94
Collaboration

The project required a lot of new research that Investment and the Financial Strategy Group
would be costly and onerous for one institution teams,80 as well as financial analysts, asset-class
to undertake independently. Collaboration brings experts and asset-allocation specialists around
the benefit of combining different geographies and the globe.
perspectives, increasing the global applicability of the
findings. Successful examples of collaboration should Outside experts
encourage institutional investors to work together,
pooling knowledge and finding solutions to market n The Grantham Research Institute on Climate
shortcomings, especially when dealing with systemic Change and the Environment at the London School
issues that investors cannot influence individually. of Economics, jointly with Vivid Economics, were
engaged as specialists on the economic impact of
This project is part of an increasing trend towards climate change, with some of the team members
pension funds and institutional investors working having been part of the Stern Review (2007). The
together in a collaborative way to address systemic team provided research input on the first two
issues that may affect their ability to meet their stages of the project – namely, designing the
long-term objectives.77 Research institutions climate scenarios and evaluating the outcome
have examined the merit of collaboration among of the scenarios on key macroeconomic and
financial professionals, fuelling further interest and micro variables. This involved the delivery of two
collaborative activities among institutional investors (non-public) reports, one on building the climate
on a range of investment issues (Guyatt, 2008; 2009).78 scenarios and the second on mapping the evidence
to the scenarios. The team also provided ongoing
The core project group support and feedback on Mercer’s interpretation of
the research outputs for investments.
n All the members of the group shared a common
goal – namely, to examine the implications of n A Research Group was established to provide
climate change for strategic asset allocation (SAA) expert input and commentary on the methodology
decision making. and draft reports.81 This group met three times in
person to debate and discuss the project outputs
n The group included 14 institutional investors during the course of the project. Interaction with
spanning Asia, Australia, Europe, the UK, the US the research group also took place outside of these
and Canada.79 Of these 14 institutional investors, a meetings, including through their involvement in
range of different types is represented with national some of the workshop events hosted during the
(wealth) funds, public pension funds (or linked to project, informal meetings, discussions and written
public sector agency), industry funds (that represent feedback on the draft reports. Group members
members for particular industries) and a company include:
pension fund.
– Alan Miller (Principal Climate Change Specialist
n The group included two industry groups with a Environment Department, IFC)
special interest in climate change and mobilising
– Dr. Monica Araya (Senior Associate, E3G, Third
private-sector finance.
Generation Environmentalism, UK)

n The project was put together and managed by – Ingrid Holmes (Programme Leader Low Carbon
Mercer, with representatives from the Responsible Finance, E3G, Third Generation Environmentalism,
UK)

77 The Principles for Responsible Investment (PRI), the Institutional Investors Group on Climate Change (IIGCC) in the UK/Europe, Ceres and the Investor Network on
Climate Risk (INCR) in the US and the Investor Group on Climate Change (IGCC) in AsiaPac, to name a few.
78 For more information, see http://www.rotman.utoronto.ca/icpm/, including sponsored research and workshops on collaboration http://www.rotman.utoronto.ca/icpm/
details.aspx?ContentID=88.
79 For more information, see http://www.mercer.com/summary.htm?idContent=1374855.
80 For more information on Mercer’s Responsible Investment team, see http://www.mercer.com/ri.
81 The Research Group included representatives from the IFC, The Carbon Trust, Oxford University Centre for the Environment, HSBC Bank, E3G – Third Generation
Environmentalism and the Institutional Investors Group on Climate Change (IIGCC).

95 Climate Change Scenarios – Implications for Strategic Asset Allocation


– Professor Gordon Clark (Halford Mackinder expertise in climate-change science and economics,
Professor of Geography, Oxford University) the finance industry (including climate change
– Nick Robins (Head of Climate Change Centre of experts and mainstream investors), as well as the
Excellence, HSBC) project group members.
– Joaquim de Lima (Global Head of Equity
Quantitative Research, HSBC) Project stages and deliverables
– Bruce Duguid (Head of Investor Relations, The
The project consisted of three major stages:
Carbon Trust)
– Garrie Lette (Chief Investment Officer, Catholic n Stage 1 – design of the climate scenarios: led by
Super, Australia) Grantham LSE/Vivid Economics, guided by Mercer
– Stephanie Pfeifer (Executive Director, Institutional with input from the project group and the research
Investor’s Group on Climate Change) group. A (non-public) report was delivered to the
project group by Grantham LSE/Vivid Economics out
Regular communication and meetings of this process.

n Mercer established an online platform, called the n Stage 2 – mapping the evidence of the climate
Connect site, which granted access to all the project scenarios to key macroeconomic and microeconomic
group members. This password-protected site gave outcomes: led by Grantham LSE/Vivid Economics,
the group the ability to interact through the portal guided by Mercer with input from the project
as well as provide access to all the reports delivered group and the research group. A (non-public) report
during the project, relevant research documents was delivered to the project group produced by
on the economic and investment impact of climate Grantham LSE/Vivid Economics out of this process.
change, and scenario analysis and factor risk for
asset allocation. n Stage 3 – evaluating the impact of the scenarios
and the evidence, as defined in Stages 1 and 2, to
n Regular conference calls took place between the consider the impact for investments across asset
project group members (typically every six weeks) to classes and regions. This stage was led by Mercer,
discuss key stages of the project and the next steps. with input from the project group and the research
During the first half of the project (October 2009 group. A series of (non-public) reports, produced
through May 2010), Grantham LSE/Vivid Economics by Mercer, on each asset class and the overall
also participated in these conference calls to discuss implications were discussed with the project group.
the draft reports that were delivered to the group.
The primary deliverables include a Public Report and a
n One-on-one interactions took place between Mercer Tailored Report for each of the asset owner partners:
and the project group members at critical points
during the project, such as establishing the goals n Public report – The aim of the public report is to
of the tailored reports and discussion about the provide a synthesis of the key findings from the
proposed methodology and preliminary findings. study, incorporating the highlights from each of the
three defined stages.
n Two in-person meetings with the project group were
held to provide the opportunity for face-to-face n Tailored reports – Each asset owner member received
interaction and exchange between the project group a confidential report specific to the organisation’s
members on the outputs and next steps. Mercer asset mix to examine the impact of the scenarios,
hosted a London event in January 2010 and the IFC including recommendations on possible actions to
hosted a Washington event in October 2010. take.

n Two workshops were hosted by Mercer as part of n Communication and outreach – The project group
the process of building the scenarios and evaluating sought to share the broad findings with the
the investment outcomes of the scenarios. The industry, such that other institutional investors
workshops were jointly facilitated by Grantham and policymakers could consider the possible
LSE/Vivid Economics and Mercer and included implications for their organisations and/or policy
representatives from the academic community with frameworks.

96
Expanding the asset-allocation toolkit

Traditional modelling approaches do not adequately n Need to be more forward looking: Climate change
capture the nature of the economic transformation requires forward-looking analysis and cannot rely
process and the potential sources of risk associated on modelling historical asset-class relationships
with climate change. As such, the tools to integrate that traditional modelling analysis techniques
climate change into the way we think about sources of predominantly rely on. This means utilising tools
risk for SAA need to be expanded along the following such as scenario analysis.
lines:
n Need to go beyond quantitative analysis: Qualitative
n Climate change increases uncertainty: Climate factors need to be embedded into the decision-
change increases the uncertainties for institutional making process. SAA decision-making processes rely
investors that can potentially have a significant heavily on quantitative analysis, whereas much of
impact on the performance of a portfolio mix the source of investment risk around climate change
over the long term, with the primary source of requires the exercise of judgement about how things
risk coming from uncertainty around policy and might develop in terms of the science of climate
its associated adjustment costs. Prudent risk change, the policymakers’ response and the type of
management processes should build climate technologies that may/may not prosper.
change considerations into long-term strategic
decision-making processes to help manage these n Need to review assumptions regarding market
uncertainties. risk: Past periods of positive and negative
economic transformation have been associated
n Need to look beyond macroeconomic impacts: The with a significant change in the realised ERP82
Grantham LSE/Vivid Economics analysis showed over time, ranging from destructive wartime
that the potential impact of climate change on GDP, periods to positive periods of substantial efficiency
interest rates and inflation across the scenarios improvements arising from a growing service sector
magnifies beyond 2050, but may not be the driving and innovations in IT. Assumptions regarding the
force behind investment risks before then. Mercer’s ERP should therefore be reviewed in light of the
analysis concluded that the source of investment potential impacts of climate change on the process
risk over the coming 20–30 years will come through of economic transformation that may take place in
the increased uncertainty around new technology, the transition to a low-carbon global economy.
physical impacts and climate policy.

n Need to think about diversification across


sources of risk: To varying degrees, traditional
asset-allocation techniques optimise portfolio
exposure based on assumptions about the risk,
return and correlation between asset classes where
diversification across assets is sought. An additional
tool in this analytical framework is to think of SAA
in terms of diversifying across sources of risk, rather
than via asset classes per se. This means utilising a
factor risk approach to supplement asset allocation
decision making.

82 Broadly defined, the ERP represents the compensation for taking on equity risk versus a risk-free rate. The notion of the ERP is widely used in finance models and also
features as an input into the way Mercer develops some of its asset-class assumptions. Hence, it is important to consider if the climate change scenarios might affect
the ERP and, if so, in what way and by how much. The ERP discussion in this study focuses on realised returns for an existing portfolio of assets at a future point in
time. This is because the study is evaluating the outcome/consequence of different climate scenarios for an existing portfolio of assets, starting from today and looking
at a future end date (in this case 2030).

97 Climate Change Scenarios – Implications for Strategic Asset Allocation


Scenario analysis

It was against this backdrop that scenario analysis While the future holds many uncertainties that
emerged as a potentially useful tool to utilise for this we may not always be able to fully prepare for, it is
project. Scenarios have been widely used, and have beneficial for long-term investors to utilise processes
proved to be a powerful tool in informing strategic that help them to better consider systemic risks such
decisions in the face of deep uncertainty about the that they respond in a more measured way if/when
future. Shell, for example, has pioneered the use of events do unfold. Scenario analysis can help fiduciaries
scenarios of future energy supply and demand, in order to fulfil their obligations in a number of ways:
to consider the risks and opportunities to its business
(Shell, 2008). Governments are also turning to scenarios n Increase knowledge and awareness about where
to understand how future events might affect areas major risks might lie across investments
of national interest. The UK government, for example,
launched the Foresight programme in 1993, in order n Become better prepared for turbulent times,
to identify risks and opportunities for the national minimising the risk of making bad (short-term)
science, technology and engineering sectors. Scenario decisions at the wrong time in response to
planning is a key element of the Foresight approach. unforeseen events

In the context of climate change, scenarios have n Recognise early warning signs if developments move
been used to map the evolution of greenhouse gas towards a certain scenario outcome
emissions, temperatures and impacts, both under
BAU and with policy intervention. Perhaps the best-
known example is the IPCC’s Special Report on Emissions
Scenarios (Nakicenovic and Swart, 2000).

Scenario analysis is ideally suited to exploring


extreme events and searching for “black swans”
(Taleb, 2007). It is “the methodical thinking of the
unthinkable” (Van der Heijden, 1996). In doing so, its
particular strength lies in identifying storylines or
sequences of events and their consequences. This can
reveal unexpected futures, but at the same time it can
also reveal inevitable futures, both of which constitute
valuable knowledge.

Not only has scenario analysis been the method


utilised by the research community on climate
change science and economics, it is also a tool that
can potentially improve risk management and SAA
processes to incorporate more qualitative factors into
the mix that are generally overlooked.

98
Factor risk framework

The next feature of our approach was to look at SAA Interpretation of the sources of investment
through a factor risk framework, as climate change risk
and scenario analysis supported a more forward-
looking approach to better inform the assumptions
Thinking about SAA in terms of the potential source
made within traditional modelling techniques that
of investment risk is a relatively new approach for
rely heavily on historical data. In its most extreme
institutional investors – hence, a few observations
form, thinking about asset allocation in terms of factor
on how a factor risk framework can be used for SAA
risks means that the decision-making framework is
decision making might be helpful at this point.
not divided up along asset-class lines but by sources
of risk. The asset classes are then thought of in terms
First, diversification across the different sources of risk
of how they will be affected by those sources of risk,
is preferable to ensure that the portfolio is resilient to a
with the ultimate goal being to achieve diversification
number of different potential factors that might affect
across them.
performance and is not overly exposed to each one.
The objective is not to maximise exposure to one factor
Scenario analysis and a factor risk framework are
or minimise it to another, but rather to have an asset
particularly helpful in considering how climate change
mix that is broadly dispersed across the sources of risk.
might affect a portfolio’s asset mix, since the sources of
Put simply, this means having a spread of high and low
risk might not always come through financial variables
exposure across the portfolio mix to fundamental
that traditional portfolio optimisation models rely on.
risks, market risks, asset-specific risks and climate
Moreover, historical data are not always available for
change risks.
risks such as climate change, since it is more about
looking into the future and less about modelling
Second, a high sensitivity to a source of risk does not
the past. It involves an element of judgement and
necessarily indicate that it will be negative (or positive)
discussion about the assumptions as part of the
for investments. For example, listed equity has a high
SAA process to test and challenge the quantitative
sensitivity to the economic cycle, but this relationship
assumptions that underpin modelling analysis.
can be either positive or negative depending on the
different stages of the economic cycle. Likewise,
Defining the sources of investment risk assets that are highly sensitive to regulatory or
political change, such as real estate and infrastructure,
The sources of risk that have been examined in this may benefit from favourable policies or suffer from
study are summarised in Table 24. The fundamental unfavourable ones. In summary, it need not be that
and market risks draw from Mercer’s analysis within a higher sensitivity to a factor means lower returns
the Growth Portfolio Toolkit (GPT).83 The climate or higher risk; the objective is to disperse the source
change risks (TIP™) are an additional set of factors of risk across the factors as far as possible and to
that have been developed for this project to provide a consider these as part of a fiduciary’s regular strategic
framework to translate the climate scenarios and review discussions.
their outcomes into sensitivities across asset classes
and regions. Finally, interpreting the direction of the potential risk
(whether it will be positive or negative) requires an
element of judgement and discussion about the
Table 24 changing investment conditions. This is one of the
Source of investment risk primary benefits of integrating factor risk analysis
into asset-allocation discussions, as it encourages
Fundamental Market factors Climate change decision makers to step back from quantitative model
factors factors
assumptions and ask questions to test their thinking.
Economic cycle Equity risk Technology (low This might be in the form of stress testing for the
sensitivity premium carbon) impact of extreme situations, or combined qualitative
and quantitative scenario analysis, the latter of which
Inflation sensitivity Volatility Impact (physical) is the method adopted for this study.
Policy (climate)
Source: Mercer

83 See ”Diversification: A Look at Risk Factors”, available at http://www.mercer.com/referencecontent.htm?idContent=1378620.

99 Climate Change Scenarios – Implications for Strategic Asset Allocation


Climate change risks – TIP™ framework

Recognising the limitations of existing approaches to way. Each factor is designed to provide insight into a
SAA and the need to integrate a plethora of new data different part of the climate change transformation
into the decision-making process around the science process. These changes may be positive or negative for
and economics of climate change, Mercer developed investments; hence, the direction of the investment
the TIP™ factor risk framework, defined as: impact is determined through interpretation of the
sensitivities of each asset class to these factors across
n Technology (T) – broadly defined as the rate of the scenarios.
progress and investment flows into technology
related to low carbon and efficiency, which are The TIP™ framework was formulated to examine the
expected to provide investment gains process of economic transformation due to climate
change, inspired by Schumpeter’s work (1947) on
n Impacts (I) – the extent to which changes to the the process of economic change and the role of
physical environment will affect (negatively) on innovation and disruptive technologies. The framework
investments is designed to look beyond the macroeconomic
impacts to understand the process of transformation
n Policy (P) – the cost of climate policy in terms of the associated with climate change and what this means
change in the cost of carbon and emissions levels for investments, such as capital flows in low-carbon
that result from policy depending on the extent to technology and energy efficiency, the sensitivity of
which it is coordinated, transparent and timely investments to physical changes to the environment,
the policy measures used and how this varies by
The factors are all interdependent, as policy will be a region, the market sensitivity to policy measures, and
key for mobilising technology, both of which will be the degree to which investors anticipate (or do not
important for minimising physical impact risk. For this anticipate) the climate policy measures.
reason, the framework cannot be viewed in a linear

Figure 16
TIP™ = Technology, Impacts and Policy
Factor risk approach to evaluate climate change investment impacts

Investment in energy efficiency, technology


development and deployment

Technology

Physical changes to our environment, Impacts Policy


Changes to carbon costs and emissions
health and food security (Physical) levels as a result of policy measures

Source: Mercer

100
As Figure 17 describes, the initial process of This combination of quantitative and qualitative
aggregating the relevant data to measure the TIP™ techniques allowed some degree of numerical
values across the scenarios was largely a quantitative estimation to gauge the possible magnitude of impacts
process. The next stage involved evaluating the and how they vary across the scenarios (such as
sensitivities of each asset class and region to the capital flows due to technology, the costs of physical
sources of investment risk, which was largely a damage to the environment and the outcome of policy
qualitative process. measures). However, the forward-looking nature of the
process also necessitated a high degree of judgement
regarding the sensitivity of each asset class and region
across the scenarios.

Figure 17
Framework linking the climate scenarios to sources of investment risk

Define sensitivities Develop and apply a T, I and P climate change risk framework to translate the climate
of asset classes to scenarios into investment impacts. Constructed via a quantitative process.
sources of risk

Consider sensitivity to
different sources of risk
across the scenarios

Qualitative evaluation of the sensitivity Finding: Higher volatility results where policy
and direction of impact for each asset uncertainty is high. Lower risk adjusted returns for
class to sources of risk late, sporadic or no climate policy scenarios.

Qualitative assessment as to whether the Finding: Infrastructure, real estate, private equity,
baseline risk/return assumptions need to sustainability and renewable energy most sensitive
change (dialogue and debate)

Interpret the results in the context of


broader asset-class characteristics and
commonalities

Discuss preliminary findings and


conclusions with all stakeholders and
experts*

Review conclusions to reflect


iterations with stakeholder Final stage: Quantitative analysis integrating sensitivities per
group climate scenario, applied to project partner’s asset mix

Source: Mercer
* Where the stakeholders and experts consulted include: All members of the project group, Mercer asset-class experts for equities, bonds, private equity, infrastructure,
real estate and factor risk specialists on asset allocation. The Research Group and Grantham LSE/Vivid Economics were also consulted.

101 Climate Change Scenarios – Implications for Strategic Asset Allocation


Figure 18 sets out the TIP™ factor risk framework. the second stage of the analysis the sensitivity of each
While each of the T, I and P factors have been asset class to these T, I and P factors is evaluated to
quantified to provide a sense of scale and magnitude determine the overall impact for investments in terms
in terms of capital flows, the framework is largely a of the magnitude and direction across the scenarios.
qualitative one that requires judgement in interpreting
the impact for investments. The T, I and P factors are all interdependent (as policy
will be key for mobilising technology, both of which
A higher T, I or P value of one scenario versus will be important for minimising physical impact
another indicates a higher value associated with the risk), hence the framework cannot be treated in a
transformation process. The variability across the linear model way. Each factor is designed to provide
scenarios is also of interest as it shows the potential insight into a different part of the climate change
range of outcomes where a higher range suggests a transformation process, some of which will be positive
higher degree of uncertainty for investors in predicting or negative for investments; hence, the direction of
the factor. Needless to say it will not always be the the impact is determined through interpretation of the
case that a higher T, I or P value will be positive for sensitivities of each asset class across the scenarios.
investments, and vice versa for lower values. Hence in

Figure 18
Climate change risks – TIP™ framework formulation

Technology: $ size of Impacts: $ cost of physical Policy: $ change in cost of


additional low carbon climate change impacts by emissions to 2030 as a result
investment flows by 2030 2030 of climate policy

Cumulative additional Cumulative economic cost Change in cost of emissions


investment in efficiency of changes to the physical = [2030 Emissions x $ /
improvements, renewable environment, health and food tCO2e] – [2010 Emissions
energy, biofuels, nuclear and security to 2030 (Source: x $ / tCO2e] (Source: CAIT
CCS to 2030 (Source: derived estimates by Grantham LSE/ and Grantham LSE/Vivid
by Mercer from IEA WEO Vivid Economics) Economics)
2009)

IEA estimates modified Calculations by Grantham Carbon price derived


according to different LSE/Vivid Economics, using by Grantham LSE/Vivid
degree of mitigation across Hope’s PAGE2002 model Economics from the WITCH
scenarios. Climate Breakdown estimates and data on model; emissions derived
is baseline investment flows adaptation costs from the by Grantham LSE/Vivid
that would happen without World Bank/United Nations Economics based on Bowen &
additional mitigation Framework Convention on Ranger, 2009 and IEA 2009
Climate Change (UNFCCC)

Result: The value of additional Result: The costs range in Result: The increase in the
investments in these assets the order of $70 bn to $180 cost of emissions from 2010
will grow by between $180 bn pa globally in terms of to 2030 ranges between
bn to $260 bn pa to 2030 for adaptation and residual $130 bn and $400 bn pa
all mitigation scenarios, with damage costs, with Climate globally, with Delayed Action
Stern Action at the upper end Breakdown the highest cost the most costly due to late
and unanticipated policy

Source: Mercer. The factors have been discounted to the net present value using a 3% discount rate. This was chosen based on a
composite of global 10Y bond yields as at October 2010.

102
Technology – Size of future investment flows Figure 19
New investment by region 2009 (total investment = $19 bn)
n This factor measures the cumulative additional
Europe
investment in low-carbon technology that takes 2% North America
place under the different climate scenarios to 2030
Latam
as a result of mitigation policies. This includes
Asia & Oceania
investment in efficiency improvements, renewable 37%
34% MENA
energy, biofuels, nuclear and CCS.

n The variability in Technology across the scenarios


has been calculated based on the expected size
10%
of the investment for each scenario, rather than 17%
on making explicit assumptions about the rates
of return for each type of technology, as the
information available is too sporadic to make any Source: Global Trends in Sustainable Energy Investment 2010 Analysis of
IRR assumptions sufficiently robust. The estimates Trends and Issues in the Financing of Renewable Energy and Energy Efficiency.
for technology therefore err on the side of caution UNEP, SEFI and Bloomberg New Energy Finance

and are likely to underestimate the future value of


technology investments, depending on the rate of n On the basis of these trends, the Regional
return earned. Divergence scenario applied a percentage reduction
in investment in some regions versus Stern Action,
n Technology estimates for Climate Breakdown are which assumes global participation and, hence,
zero, as the data are based on the IEA World Energy represents the best-case outcome in terms of
Outlook (2009) computations on the additional investments. For Regional Divergence:
investment under the 450 Scenario used as the
proxy for Stern Action versus the Reference baseline – The leaders were assumed to be 100% of the Stern
scenario as the proxy for Climate Breakdown. In Action investment levels. This applied to the EU
Regional Divergence, we have assumed 25% less and China/East Asia.
investment spending than Stern Action, as emissions – The mature but contracting was assumed to be
levels are higher and policy is less efficient globally. 50% of the Stern Action investment levels. This
Delayed Action only includes the 2020–2030 of the applied to the US.
IEA estimates of investment flows, as we assume – The improvers were assumed to be 50% of the
BAU until after 2020. Stern Action investment levels. This applied to
India/South Asia and Japan.
n By region, the estimates are also based on the
– The laggards were assumed to be 20% of the Stern
IEA World Energy Outlook (2009) data together
Action investment levels. This applied to Russia.
with Mercer assumptions. The same estimation
process as defined above was applied, with the only
modification being a clustering of countries into
leaders (the EU and China/East Asia), mature but
contracting (the US), improvers (Japan and India/
South Asia) and laggards (Russia). This classification
was based on the rate of change in the investment
into clean energy and energy efficiency, as reported
by UNEP, SEFI and Bloomberg New Energy Finance
(2010). Figure 19 summarises the new investment
flows by region in 2009, with China representing a
substantial $33.7 bn and India $2.7 bn of the Asia
and Oceania category.

103 Climate Change Scenarios – Implications for Strategic Asset Allocation


Impacts – Cost of physical climate change (2009a) study, which takes a sectoral approach to
impacts estimating adaptation costs but provides estimates
disaggregated to the level of large regions in the
n This factor measures the economic cost of changes developing world, and shows how these change over
to the physical environment, health and food time, implicitly as a function of climate change and
security owing to climate change. It was estimated socioeconomic development. Appendix F provides
by Grantham LSE/Vivid Economics on the basis of further explanation of the methodology used by
two components – adaptation costs and residual Grantham LSE/Vivid Economics.
damages. The impact on adaptation costs and
residual damage in terms of GDP effect over the n As Table 25 shows, the estimated difference between
scenarios out to 2030 is presented in Table 25. The adaptation costs between the Climate Breakdown
costs of adapting to climate change include costs and Stern Action scenarios across the countries
related to infrastructure, coastal zone protection, is low in terms of percentage GDP by 2030 (the
extreme weather, human health, fisheries, focus of this study). However, the differences in
agriculture and water supply/flood protection. adaptation costs between the scenarios become
Adaptation costs also incorporate the financial more pronounced over time. This is because the
transfers required to assist adaptation from change in temperature and sea-level rise escalates
developed to developing economics. by 2050 and beyond. The focus of this study is based
on the estimates to 2030 – hence, the differences in
n Adaptation costs were estimated by Grantham temperature are comparatively small (as illustrated
LSE/Vivid Economics based on a World Bank in Figure 20 on page 105).

Table 25
Adaptation costs are higher under Climate Breakdown, and increase more than proportionately over time

Adaptation costs in US$ billion and % GDP (in parenthesis)

2012 2020 2030 2050

Climate Breakdown

Europe 7.3 (0.1) 10.5 (0.1) 17.1 (0.1) 38.4 (0.2)

US and Canada 27.4 (0.3) 39.3 (0.3) 63.7 (0.4) 143.5 (0.7)

OECD Pacific 3.5 (0.0) 5.0 (0.1) 8.1 (0.1) 18.2 (0.1)

China and East Asia 22.3 (0.4) 23.6 (0.2) 25.7 (0.1) 32.2 (0.1)

Russia and the former 6.4 (0.5) 8.3 (0.4) 11.6 (0.3) 22.2 (0.3)
Soviet Union

Latin America and the 18.4 (0.7) 19.6 (0.5) 21.8 (0.4) 28.1 (0.3)
Caribbean

Middle East and 1.8 (0.1) 2.8 (0.1) 4.4 (0.1) 9.7 (0.1)
North Africa

India and South Asia 10.0 (0.4) 11.3 (0.4) 13.6 (0.2) 20.7 (0.1)

Sub-Saharan Africa 12.6 (0.7) 15.7 (0.6) 21.2 (0.8) 38.7 (1.0)

104
Table 25
Adaptation costs are higher under Stern Action, and increase more than proportionately over time (cont’d)

Adaptation costs in US$ billion and % GDP (in parenthesis)

2012 2020 2030 2050

Stern Action

Europe 7.3 (0.1) 9.2 (0.1) 11.5 (0.1) 17.1 (0.1)

USA and Canada 27.4 (0.3) 34.5 (0.3) 43.0 (0.3) 63.7 (0.3)

OECD Pacific 3.5 (0.0) 4.4 (0.0) 5.4 (0.0) 8.1 (0.1)

China and East Asia 22.3 (0.4) 23.2 (0.2) 24.3 (0.1) 26.1 (0.1)

Russia and the former 6.4 (0.5) 7.4 (0.3) 8.5 (0.3) 11.1 (0.2)
Soviet Union

Latin America and the 18.4 (0.7) 19.4 (0.5) 20.5 (0.4) 22.5 (0.2)
Caribbean

Middle East and 1.8 (0.1) 2.3 (0.1) 2.9 (0.1) 4.2 (0.1)
North Africa

India and South Asia 10.0 (0.4) 10.7 (0.2) 11.5 (0.1) 13.2 (0.1)

Sub-Saharan Africa 12.6 (0.7) 14.2 (0.6) 16.2 (0.6) 20.6 (0.5)
Source: Grantham LSE/Vivid Economics calculations, based on World Bank (2009a)

Figure 20
Rapid warming under Climate Breakdown, lower under Stern Action
Climate Breakdown

Stern Action

3
Global mean temperature rise above

2.5
pre-industrial (ºC)

1.5

0.5

0
2010 2020 2030 2040 2050
Source: Grantham LSE/Vivid Economics

105 Climate Change Scenarios – Implications for Strategic Asset Allocation


n Another variable that could change between integrated assessment models, PAGE2002 contains
the scenarios by 2030 is residual damage costs. a set of equations to represent all links in the
Adaptation will not eliminate all of the damage chain between economic and population growth
costs of climate change (Fankhauser, 2010). Even if it and associated greenhouse gas emissions on one
were technically feasible to do so, it would be highly side and economic damages of climate change
unlikely to be cost-effective, as adaptation costs at on the other. The section “Estimating Residual
the margin are likely to exceed damage costs when Damages” (on page 11) provides further explanation
near total climate proofing is achieved. Therefore, of the methodology used by Grantham LSE/Vivid
the residual damage costs of climate change must Economics.
still be determined.
n Table 26 estimates residual damage costs between
n The residual damage cost estimates were based the Stern Action and Climate Breakdown scenarios.
on the PAGE2002 model that was built by Dr. Chris Residual damages are higher under Climate
Hope from the University of Cambridge and used Breakdown than under Stern Action – even by 2030.
by the Stern Review (Stern, 2007). It captures sub- This is because of the temperature differences
factors such as changes in temperature, increased between the two scenarios (see Figure 20) as well
propensity for drought, increased flood risk, as using PAGE2002 to assume that a given amount
increased risk of extreme weather events, changes of warming produced more damage in Climate
in health and disease across regions, and changes in Breakdown than in Stern Action to reflect the
crop yields and food security. longer-term policy trajectory. Furthermore, it is again
the case that residual damage costs increase over
n PAGE2002 has eight world regions and a time time – hence, this difference between the scenarios
horizon of 200 years, from 2000 to 2200. Like other increases considerably by 2050 and beyond.

Table 26
Residual climate damages are steeply rising under Climate Breakdown and are highest in the developing world

Residual damage costs in US$ billion and % GDP (in parentheses)

2012 2020 2030 2050

Climate Breakdown

Europe 0.0 (0.0) 0.0 (0.0) 1.3 (0.0) 10.3 (0.1)

USA and Canada 0.0 (0.0) 0.0 (0.0) 0.7 (0.0) 5.9 (0.0)

OECD Pacific 0.1 (0.0) 0.0 (0.0) 1.9 (0.0) 4.8 (0.0)

China and East Asia 0.0 (0.0) 0.1 (0.0) 4.4 (0.0) 44.0 (0.1)

Russia and the former 0.0 (0.0) 0.0 (0.0) 0.1 (0.0) 1.2 (0.0)
Soviet Union

Latin America and the 6.1 (0.2) 20.3 (0.5) 70.1 (1.2) 285.4 (2.8)
Caribbean

Middle East and 6.3 (0.3) 20.9 (0.6) 45.8 (1.3) 154.2 (3.1)
North Africa

India and South Asia 4.0 (0.2) 12.5 (0.3) 57.4 (0.7) 288.2 (1.5)

Sub-Saharan Africa 4.8 (0.3) 16.1 (0.3) 35.3 (1.3) 116.4 (3.1)

106
Table 26
Residual climate damages are steeply rising under Stern Action and are highest in the developing world (cont’d)

Residual damage costs in US$ billion and % GDP (in parentheses)

2012 2020 2030 2050

Stern Action

Europe 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) 0.4 (0.0)

USA and Canada 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) 0.1 (0.0)

OECD Pacific 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) 0.3 (0.0)

China and East Asia 0.0 (0.0) 0.0 (0.0) 0.2 (0.0) 2.0 (0.0)

Russia and the former 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) -0.1 (0.0)
Soviet Union

Latin America and the 1.1 (0.0) 2.9 (0.1) 8.7 (0.2) 33.0 (0.3)
Caribbean

Middle East and 2.1(0.1) 8.1 (0.2) 5.7 (0.2) 17.9 (0.4)
North Africa

India and South Asia 0.8 (0.0) 2.1 (0.0) 7.6 (0.1) 33.8 (0.2)

Sub-Saharan Africa 1.6 (0.1) 6.2 (0.2) 4.4 (0.2) 13.5 (0.4)
Source: Grantham Research Institute/Vivid Economics

n On the basis of these estimates for Climate Figure 21


Breakdown and Stern Action, the following Residual damage costs + Adaptation costs as a percentage GDP
assumptions were made for the remaining to 2030
scenarios:
Residual damage

– For adaptation costs, as Figure 21 shows, both Adaptation


scenarios are the same as for Climate Breakdown
and Stern Action at around 0.2% of global GDP in 0.5
2030. This is because policy cannot change the
course of the physical impacts of climate change
0.4
within that period.
– For residual damage costs, based on the emissions
trajectory for each scenario, the Delayed Action 0.3
% GDP

scenario was assumed to cost 1.5x that of Stern


Action in terms of residual damage. Similarly,
0.2
Regional Divergence was assumed to be 2x the
cost of Stern Action. This is because the residual
damage cost estimates take into account the 0.1
emissions trajectory, with Regional Divergence
producing higher emissions levels than under
Delayed Action (see Table 27, on page 108, for 0.0
emissions levels for each scenario). Regional Delayed Stern Climate
Divergence Action Action Breakdown

Source: Grantham LSE/Vivid Economics

107 Climate Change Scenarios – Implications for Strategic Asset Allocation


Policy – Change in cost of emissions Table 27
Cost of carbon and emissions levels in 2030
n This factor measures the change in the cost of
Scenario Cost of carbon in 2030 Emissions in 2030
carbon emissions as a result of climate policy
measures in 2030 under each scenario. It is derived Regional $110/tCO2e only in 50 Gt CO2e per
from the change in the expected level of emissions Divergence participating regions, year
and the future carbon price in 2030 compared to including all countries in
2010. More specifically, it represents the change in this study except India/
South Asia and Russia
future emissions level x future carbon price between
the two periods. The carbon price could be a market- Delayed $15/tCO2e to 2020 then 40 Gt CO2e per
based trading system or an implied cost of carbon Action dramatic rise to $220/ year
due to policy measures and/or relative changes in tCO2e
commodity prices. Stern $110/tCO2e 30 Gt CO2e per
Action year
n The estimates are based on the outcomes of the
Climate $15/tCO2e 63 Gt CO2e per
climate scenarios within the Grantham LSE/Vivid
Breakdown year
Economics scenarios and mapping evidence reports
produced as part of this project. The assumptions Source: Grantham LSE/Vivid Economics estimates

are set out in Table 27. In brief:


n Across regions, the same methodology and source
– The carbon price estimates were derived by of data were used for the regions examined in this
Grantham LSE/Vivid Economics from the WITCH study for each scenario, with the future carbon
model of the RECIPE study, built by a team from emissions in each respective region/country in 2030
the FEEM research institution. using IEA WEO (2009) data for the 450 Scenario as a
– The emissions levels were derived by Grantham proxy for Stern Action and the Reference scenario
LSE/Vivid Economics based on data from Bowen for Climate Breakdown. The Delayed Action scenario
and Ranger (2009) and the IEA (2009), in which the was estimated based on a shortfall in reducing
450 Scenario was used as a proxy for Stern emissions at the global level as a result of the delay
Action and the Reference scenario as a proxy for compared to Stern Action – that is, emissions stay
Climate Breakdown. higher by a factor of 40 Gt/30 Gt (see Table 27). The
Regional Divergence scenario applied the same
n According to Grantham LSE/Vivid Economics, the factor weighting based on comparative emissions
carbon price estimates are in the middle of model levels to Stern Action of 50 Gt/30 Gt.
estimates in the RECIPE study over the time period,
but increase rapidly after 2030. Prices accelerate n The cost of carbon for the non-participating
because the model’s agents have perfect foresight – countries in the Regional Divergence scenario –
they require a relatively modest carbon price to take namely Russia and India/South Asia – was assumed
early action (given their expectation of higher rises to be $15/tCO2e. The other countries examined
subsequently), but they expect limited technological in this report that are assumed to incur a cost
options and substitution possibilities within the of carbon of $110/tCO2e include the EU, the US,
energy sector later, because cheaper options are China/East Asia and Japan. For all other scenarios
exhausted earlier on. Nearly all projections of carbon the carbon cost was applied consistently across
prices entail period-by-period increases for several the countries and analysed against their future
years, often well into the second half of the century emissions levels.
or beyond.

108
Estimating the costs
of adaptation
This section was produced by Grantham
LSE/Vivid Economics.

109 Climate Change Scenarios – Implications for Strategic Asset Allocation


Estimates from the literature of the costs of adaptation, Given this assumption, the constants αr,s and βr,s can
in particular from the World Bank (2009) and from the be solved by simultaneous equations to calibrate the
United Nations Framework Convention on Climate curve on data from the World Bank (2009) study. The
Change (UNFCCC) (2007), are reported with respect to resulting adaptation cost curve thus delivers the same
a particular scenario of greenhouse gas emissions and estimate of the cost of adaptation as the World Bank
associated climate change. The UNFCCC estimates (2009) study using its warming scenario, but it can also
have the added limitation of only being available for be extrapolated and interpolated to different amounts
one future year, 2030. In their case, the dependency of of warming.
costs on climate change cannot be deduced.
Adaptation cost curves are constructed for three
In order to produce estimates of adaptation costs that sectors: (i) infrastructure, (ii) coastal zone protection
are both dynamic and consistent with our scenarios, and (iii) all other adaptation costs. The third category
a series of adaptation cost functions were constructed includes the following sectors: industrial and
that show how the costs of adaptation in particular municipal water supply and riverine flood protection,
regions and sectors depend on a changing climate. agriculture, fisheries, human health, and extreme
events. There is insufficient information in the
The following functional form was used: World Bank (2009) study to estimate adaptation cost
functions for these sectors separately. However, data
are available on the share of these sectors’ costs in (iii),
2
Cr , s ,t  α r , s  β r , sTEt (AX1) averaged over the period 2010–2050, and so a rough
estimate of costs to these sectors can be made by
assuming they command a constant share of (iii) for all
Where C is the cost of adaptation in region r, sector s, amounts of warming.
and at time t, αr,s and βr,s are constants, and TE is the
increase in global mean temperature above the pre-
industrial level.

One exception is the cost of enhancing coastal zone


protection, in which costs were set as a function of
sea-level rise directly:

2
Cr , s ,t  α r , s  β r , s SLRt (AX2)

Where SLR is sea-level rise.

Costs are assumed to be a quadratic function of


warming/sea-level rise, meaning that they increase
more than proportionately. That costs increase more
than proportionately with increasing adaptation
requirements is a common finding, while the quadratic
functional form is convenient and is frequently used
in the economics of climate change to represent cost
functions of all types.

110
Estimating residual
damages
This section was produced by Grantham
LSE/Vivid Economics.

111 Climate Change Scenarios – Implications for Strategic Asset Allocation


To estimate the residual damages of climate change, Where D is residual damage in region r at time t, TE
the PAGE2002 model was used. This model was built is the increase in regional mean temperature (which
by Chris Hope (2006) and used by the Stern Review PAGE2002 estimates from the corresponding increase
(Stern, 2007). PAGE2002 has eight world regions and in global mean temperature), α sets the cost of 2.5°C
a time horizon of 200 years, from 2000 to 2200. Like warming, and β is the damage-function exponent,
other integrated assessment models (IAMs), PAGE2002 determining curvature. Adaptation reduces the cost of
contains a set of equations to represent all links in the climate change in two ways. First, it sets the tolerable
chain between economic and population growth and level of warming before any residual damage occurs,
associated greenhouse gas emissions on one side and ATL. Second, it reduces damage in excess of that
economic damages of climate change on the other. tolerable level by the amount K. ATL is itself a function
Figure A1 (on page 113) is a stylistic representation of both of the level of warming ATP and the rate of
this process. warming ATR:

PAGE2002 is a stochastic IAM, meaning that each of


its parameters can be sampled from a distribution ATLr ,t  min[ ATPr ,t , ATLr ,t −1  ATRr ,t t )] (AX2)
of values in the course of a Monte Carlo simulation,
which ultimately produces a distribution of estimates
of the cost of climate change (and, if required, of the For warming of up to around 3°C, the key parameter
cost of adaptation and of mitigation). However, since is α, the regional cost of 2.5°C warming. Under the
the present study is scenario-based, the model was Climate Breakdown scenario, it is assumed that αr
run in deterministic mode with a set of assumptions take their maximum value, which is consistent with
consistent with each scenario. the scenario’s storyline. In Stern Action, it is assumed
that they take their central value. In both scenarios,
Most of the intricacies of the modelling need not be the curvature of the damage function β is set to the
explored in detail, since not all aspects of the model central value of 1.8 (that is, almost quadratic). β only
were used in this study. Rather, the question is: becomes an important parameter for warming well in
What are the residual damages of climate change for excess of 3°C.
scenarios of global mean warming that are compatible
with the scenarios? The equations representing how
global mean temperature responds to increases in the
atmospheric concentration of greenhouse gases are
particularly complex. In short, the so-called climate
sensitivity parameter was calibrated – namely, the
equilibrium increase in global mean temperature
following a doubling in the atmospheric concentration
of CO2 – in order to replicate the temperature trend in
Climate Breakdown and Stern Action.

From here, the key assumptions about residual climate


damages concern the so-called damage function,
which in PAGE2002 is represented by the following
equation:

β
 TE − ATL r ,t 
Dr ,t  α r  r ,t  κ r ,t (AX1)
 2.5 

112
Figure A1
Integrated assessment models capture the whole process of man-made climate change

Baseline socioeconomic development

Economy Population Technology

Mitigation (that is, intervention to reduce emissions)

Emissions

Historical emissions of Physical change


greenhouse gases, – Carbon cycle
and natural influences
on climate (for example, – Radiative forcing
solar activity) – Global climate change
– Regional climate change
– Regional weather
– Environmental change

Adaptation

Socioeconomic impact

113 Climate Change Scenarios – Implications for Strategic Asset Allocation


114
Table listing

Tables

Highlights Table 1: Key features and potential 10 Table 13: Real estate opportunities 60
outcomes of the climate scenarios to 2030
Table 14: Real estate risks 61
Highlights Table 2: Impact of scenarios on 11
source of investment risks Table 15: Sensitivity of core unlisted real 62
estate to the TIP™ factor risks for each
Highlights Table 3: TIP™ factor risk sensitivity 15 scenario
and direction of impact for asset classes
Table 16: Sensitivities of timberland, 66–67
Highlights Table 4: TIP™ factor risk sensitivity 18 agriculture land and carbon to the TIP™ factor
and direction of impact for regions risks for each scenario

Table 1: Impact of scenarios on sources of 24 Table 17: Sensitivities of the EU, US and Japan 71–72
investment risk to the TIP™ factor risks for each scenario

Table 2: Assets with high or very high 25 Table 18: Sensitivities of China/East Asia, 73–74
sensitivity to investment risks Russia and India/South Asia to the TIP™
factor risks for each scenario
Table 3: Determinants of the ERP across the 27
climate scenarios Table 19: Uncertainty and macroeconomic 78
impact
Table 4: Value of the TIP™ factors across the 29
climate scenarios Table 20: Commodity price impact to 2030 81

Table 5: Sensitivity of asset classes to climate 33 Table 21: Energy- and carbon-intensive 84
change risks primary and manufacturing industries
would suffer the biggest impact of a carbon
Table 6: Sensitivity of regions to climate 34 constraint
change risks
Table 22: Initial climate change has both 85
Table 7: Sensitivities of global equities, 45–47
positive and negative consequences for
emerging-market equities, broad economic sectors
sustainability-themed equities and
renewables to the TIP™ factor risks Table 23: Little difference in sea-level rises 90
until after 2030
Table 8: Sensitivities of government bonds, 49–50
emerging debt, investment grade and credit Table 24: Source of investment risk 99
to the TIP™ factor risks
Table 25: Adaptation costs are higher under 104–105
Table 9: Sensitivities of private equity LBO, 52–53 Climate Breakdown, and increase more than
venture capital and renewable energy to the proportionately over time
TIP™ factor risks for each scenario
Table 26: Residual climate damages are 106–107
Table 10: Infrastructure types by sector 55 steeply rising under Climate Breakdown and
are highest in the developing world
Table 11: Sensitivities of infrastructure core 56–57
unlisted and renewables unlisted to the TIP™ Table 27: Cost of carbon and emissions levels 108
factor risks for each scenario in 2030

Table 12: Real estate and climate change 59


impacts

115 Climate Change Scenarios – Implications for Strategic Asset Allocation


Figures

Highlights Figure 1: TIP™=Technology, 8 Figure 11: Energy efficiency accounts 82


Impacts and Policy Factor risk approach to for at least half of potential abatement
evaluate climate change investment impacts
Figure 12: Governments push 83
Highlights Figure 2: Renewables and 12 innovation of new technologies,
nuclear overtake fossil fuels, in Stern Action markets pull them
scenario, by 2050
Figure 13: The carbon market already 86
Highlights Figure 3: Adaptation costs in 2030 13 has a global reach
for Climate Breakdown scenario
Figure 14: Regional temperature change 89
Highlights Figure 4: Climate change risks – 14 in 2020 (top) and 2050 (bottom) for BAU
TIP™ framework formulation
Figure 15: Percentage change in crop 91
Highlights Figure 5: Contribution to risk for 16 yield in 2020 and 2050 under climate
representative portfolio mix breakdown

Highlights Figure 6: Portfolio to target 7% 17 Figure 16: TIP™ = Technology, Impacts 100
(nominal) return and Policy Factor risk approach to evaluate
climate change investment impacts
Figure 1 : Cumulative Investment in 30
Technology to 2030 by region Figure 17: Framework linking the 101
climate scenarios to sources of
Figure 2 : Adaptation and residual damage 31 investment risk
costs by region
Figure 18: Climate change risks – TIP™ 102
Figure 3: Policy cost comparison 32 framework formulation
Figure 4: Contribution to risk for 36 Figure 19: New investment by region 103
representative portfolio mix in ‘default’ 2009 (total investment = $19 bn)
case
Figure 20: Rapid warming under 105
Figure 5: Example of portfolio mix across the 38 Climate Breakdown, lower under Stern
scenarios – portfolio to target 7% return Action
Figure 6: Delayed Action – impact of new 39 Figure 21: Residual damage costs + 107
portfolio mix on contribution to risk Adaptation costs as a percentage GDP
to 2030
Figure 7: Cost of carbon adjustment by 43
sector Figure A1: Integrated assessment 113
models capture the whole process of
Figure 8: Exposure of EM equity index to 44
man-made climate change
climate ‘impact’ risks

Figure 9: Exposure of EM bond index to 48


climate ‘impact’ risks

Figure 10: The full and long-run economic 79


cost of climate change

116
Glossary

Abatement
Mitigation cost estimates generally comprise abatement costs: the extra resources needed to produce the existing
pattern of production, sometimes with an estimate added on for the costs of reducing energy demand.

Adaptation
The process or outcome of a process that leads to a reduction in harm or risk of harm or realisation of benefits
associated with climate variability and climate change. Adaptation costs are the added costs of adapting
economies to climate change.

Alpha
Excess return to the market return, added by an investment manager through active management. It is often
referred to as manager skill.

Asset/liability modelling (ALM)


Projection of future movements in assets and liabilities, and especially, the relationship between the two. ALM
is used to provide an insight into the likely effect of different asset-allocation strategies on a pension scheme’s
future financial position.

Beta
At the asset-class level, beta refers to market return, or the return that would be earned on an asset class
independent of manager skill. In its purest form, beta can be obtained through passive investment against an
index where this is available (for example, equities, bonds, real estate, infrastructure and some commodities). For
assets that embed a component of active management, the separation of alpha (manager skill) from beta (market
return) is less straightforward.

Business as usual (BAU)


In the context of climate change and the climate change scenarios laid out in the report, BAU refers to unchanged
policy from the current situation. A BAU scenario has been an integral part of most studies into the consequences
of climate change. The most well-known and well-used set of BAU scenarios was developed by IPCC in its Special
Report on Emissions Scenarios. The Reference scenario (that is, business as usual) of the IEA has also been widely
used.

Capital asset pricing model (CAPM)


Economic model for valuing assets. The simplest version states that the expected excess return of a security over
a risk-free asset will be exactly in proportion to its beta.

Carbon capture and storage


Carbon capture and storage (CCS), alternatively referred to as carbon capture and sequestration, is a means of
mitigating the contribution of fossil fuel emissions to global warming, based on capturing carbon dioxide (CO2)
from large point sources such as fossil fuel power plants, and storing it in such a way that it does not enter the
atmosphere.

Clean energy/cleantech
Products, services and processes that are geared towards reducing or eliminating the environmental impact of
a means of production. It may include investments in agriculture, energy, manufacturing, materials, technology,
transportation and water.

Climate change
A change of climate that is attributed directly or indirectly to human activity that alters the composition of the
global atmosphere and that is in addition to natural climate variability observed over comparable time periods.

117 Climate Change Scenarios – Implications for Strategic Asset Allocation


Climate change risk factors
Climate change risk factors have been defined in this report in terms of the TIP™ framework – Technology,
Impacts and Policy. They can be used to examine the extent to which asset-class returns are sensitive to
climate change.

Climate sensitive assets


Climate sensitive assets refer to assets whose underlying risk/return characteristics are sensitive to the
different sources of risk, defined in this study as low-carbon technology (T), physical impact risk (I) and climate
policy risk (P). As indicated in Table 2, page 25, we conclude that the assets that are either highly sensitive
or very highly sensitive to climate change include real estate, infrastructure, private equity, sustainable equities
(listed and unlisted), efficiency/renewables (listed and unlisted) and commodities (including agricultural land
and timberland).

Debt/equity ratio
A company’s borrowings divided by its issued share capital. It is a measure of the amount of gearing (leverage)
of a company and an indicator of financial strength. A company with a higher debt/equity ratio can offer greater
returns to shareholders, but these will be more volatile than if the gearing were lower.

Efficiency/renewables listed/unlisted assets


In this report, this term is used to capture listed/unlisted sustainability-themed assets whose core activities are
theme-specific and more concentrated in terms of exposure than broad sustainability equity. This includes (but
is not limited to) energy efficiency, low energy transport, renewable energy, bioenergy, carbon capture and
storage, smart grid, water supply, usage and management, waste management, hydro energy and geothermal, to
name a few.

Equity risk premium (ERP)


Broadly defined, the ERP represents the compensation for taking on equity risk versus a risk-free rate. There
are different ways to measure and refer to the ERP (Fernández, 2010), including historical equity premium,
which measures the historical differential return of the stock market over treasuries. Expected equity premium
measures the expected differential return of the stock market over treasuries. Required equity premium
measures the incremental return of a diversified portfolio (the market) over the risk-free rate required by an
investor. It is used for calculating the required return to equity. Implied equity premium measures the required
equity premium that arises from assuming that the market price is correct.

Factor risk framework


A risk management tool that can complement traditional asset-allocation techniques, with the aim being to
diversify across different sources of risk and return drivers across assets. The analysis can include factors such as
the equity risk premium, small cap premium, unexpected inflation, term premium and credit risk premium. The
approach allows for scenario analysis and the inclusion of additional sources of risk that might not otherwise be
considered, such as leverage, illiquidity and climate change risk factors.

In its most extreme form, thinking about asset allocation in terms of factor risks means that the decision-making
framework is not divided up along asset-class lines but by “sources of risk”. The asset classes are then thought
of in terms of how they will be affected by those sources of risk, with the ultimate goal to achieve diversification
across them. This is the philosophy underpinning Mercer’s GPT that has been applied in this report.

Fundamental analysis
Assessment of a company’s share value and potential for future cash flows, profit and dividends based on
accounting, economic and business information (hence, fundamental factors).

Fundamental risk factors


Fundamental risk factors refer to changes in the macroeconomic backdrop that might have an impact on
investment performance, including changes to interest rates, inflation and GDP growth.

118
Impact risk factor
In this study, the impact risk factor refers to the extent to which changes to the physical environment will have
an impact (negative) on investments, representing the I of the TIP™ framework.

Long-horizon assets
Similar characteristics to real assets (defined below), in that they tend to be priced with a long-term horizon
(>10-year horizon), such as unlisted infrastructure and real estate, and are illiquid assets that cannot be
readily realised.

Market risk factors


Market risk factors refer to the broad market conditions that might have an impact on investment performance,
including factors such as the ERP, market volatility and illiquidity risk. It is the level of risk in the market that
cannot be fully eliminated by diversification. Also known as systematic risk.

Mean-variance analysis
The process of portfolio selection that assumes that every rational investor, at a given level of risk, will accept
only the largest expected return. More specifically, mean-variance analysis attempts to measure risk, correlation
and expected return mathematically to help the investor find a portfolio with the maximum return for the
minimum amount of risk. It is widely used in finance but also has a number of shortcomings, particularly the
assumptions that investors are rational, that correlations are fixed and constant and that returns are normally
distributed.

Mercer’s Growth Portfolio Toolkit


Refer to “Factor Risk Framework”.

Mitigation
Mitigation of climate change involves actions that are designed to limit the amount of long-term climate change.
Mitigation costs are the added costs of reducing greenhouse gas emissions.

Modern portfolio theory (MPT)


Modern portfolio theory is a theory of investment that attempts to maximize portfolio expected return for a
given amount of portfolio risk, or equivalently minimise risk for a given level of expected return, by carefully
choosing the proportions of various assets.

PAGE2002 model
To estimate the residual damages of climate change, Grantham LSE/Vivid Economics used the PAGE2002
model built by Chris Hope (2006) and used by the Stern Review. PAGE2002 has eight world regions and a time
horizon of 200 years, from 2000 to 2200. Like other integrated assessment models (IAMs), PAGE2002 contains a
set of equations to represent all links in the chain between economic and population growth and associated
greenhouse gas emissions on one side and economic damages of climate change on the other.

Policy risk factor


In this study, policy risk is defined as the P of the TIP™ framework, meaning the cost of climate policy in terms of
the change in the cost of carbon and emissions levels that result from policy, depending on the extent to which it
is coordinated, transparent and timely.

Portfolio risk
Represents the aggregation of risks associated with the assets held in a portfolio. In traditional approaches to
mean variance analysis, risk refers to the standard deviation in returns. In this study, the different sources of risk
underlying different assets have been examined, including fundamental risks, markets risks and climate change
risks. Portfolio risk should be considered in the context of an institution’s strategic objectives and the risk of not

119 Climate Change Scenarios – Implications for Strategic Asset Allocation


meeting these. For example, the objective may be to generate sufficient returns to hedge liabilities, to protect a
reserve pool of assets while minimising risk and maximising return, to minimise variations in contribution for
sponsors, or to target a certain funding level.

Real assets
Physical/tangible assets such as infrastructure, private equity, real estate, gold, agricultural land, timberland.

Scenario analysis
Analysis of alternative future possibilities as an input into future planning, strategic risk management and, in the
context of this study, asset allocation. Scenarios have been widely used and have proved to be a powerful tool in
informing strategic decisions in the face of deep uncertainty about the future. In the context of climate change,
scenarios have been used to map the evolution of greenhouse gas emissions, temperatures and impacts, both
under BAU and with policy intervention. Scenario analysis can help to reveal unexpected futures, but at the same
time it can also reveal inevitable futures, both of which constitute valuable knowledge.

Short-horizon assets
Liquid assets that are readily tradeable, such as stocks and bonds, that tend to be priced with a relatively short
time horizon (12–18 months) compared to long-horizon assets.

Strategic asset allocation


Broadly defined as the use of optimisation tools to determine long-term asset allocation benchmarks to
achieve long-term objectives. The objectives vary, depending on the type of asset owner and its obligations to
beneficiaries or other stakeholders. It involves making decisions around allocation to high-level asset classes –
that is, equity/fixed split, domestic/international/emerging equity split, duration of fixed income, and the split
between nominal and inflation-adjusted fixed income, allocation to unlisted assets and sustainability-themed
assets. This is distinct from other considerations such as portfolio structuring (including allocation to capital
weightings, styles and sectors, and includes active/passive analysis) and manager selection (the evaluation of
manager performance in order to select one suitable for a client’s requirements).

Sustainable investment
Broadly speaking, sustainable investment refers to investments that integrate long-term sustainability issues into
core investment-making processes. At its broadest level, sustainable investment seeks to support sustainable
economic development, enhance quality of life and safeguard the environment.

Sustainable equity
Sustainable equity refers to broad multi-themed listed equity companies that generate a substantial proportion
(typically more than 25%) of their earnings through sustainable activities. Sustainable activities at the
broadest level are those that seek to support sustainable economic development, enhancing quality of life and
safeguarding the environment.

Technology risk factor


In this study, the technology risk factor is defined as the T of the TIP™ framework, meaning the rate of progress
and investment flows into technology related to low-carbon and energy efficiency.

WITCH model
The WITCH (World Induced Technical Change Hybrid) model is one of the main modelling tools developed within
the Climate Change Modelling and Policy Research Programme of the Fondazione Eni Enrico Mattei. Grantham
LSE/Vivid Economics used this model to estimate the macroeconomic impacts of mitigation costs, adaptation
costs and residual damage costs for the Stern Action and Climate Breakdown scenarios, describing it as a “top
down” model that has considerable technological detail.

120
Bibliography

Aulisi A., Sauer A. and Wellington F. Trees in the Greenhouse: Why Climate Change Is Transforming the Forest Products
Business, World Resource Institute, 2008.

Bain & Co. Global Private Equity Report, Bain and Co., 2010.

BBC Climate Change Poll – February 2010, BBC, 2010.

Bosello F. and Roson R. “Economy-wide Estimates of the Implications of Climate Change: Human Health”,
Ecological Economics, vol. 58, no. 3 (2006), pp. 579–591.

Brenkert M., Malone A. and Moss R. (2005) as quoted in Yohe G. et al. “Global Source: Vulnerability to Climate
Change”, The Integrated Assessment Journal, vol. 6, no. 3 (2006), pp. 35–44.

Brinson G., Hood R., Beebower G. “Determinants of Portfolio Performance”, Financial Analysts Journal (July/August
1986), pp. 39–44.

Brinson G., Singer B. and Beebower G. “Determinants of Portfolio Performance II: An Update”, Financial Analysts
Journal, vol. 47, no. 3 (1991), pp. 40–48.

Brown O. “Migration and Climate Change”, International Organisation for Migration Research Series, Paper No. 21
(2008).

Damodaran A. “Equity Risk Premiums (ERP): Determinants, Estimation and Implications”, Stern School of
Business, September 2008, with an October update reflecting the market crisis.

DB Climate Change Advisors Global Climate Change Policy Tracker: An Investor’s Assessment, New York and London:
DB Climate Change Advisors, 2009.

Dimson E., Marsh P. and Staunton M. “Global evidence on the equity risk premium”, Journal of Applied Corporate
Finance, Vol. 15, No. 4 (Summer, 2003), pp. 8–19.

Edenhofer O. et al. The Economics of Decarbonization, Report of the RECIPE Project, Potsdam Institute for Climate
Impact Research, 2009.

FAO “Climate change adaptation and mitigation: challenges and opportunities for food security”, Background
report by the Food and Agriculture Organisation of the United Nations for High-Level Conference on World Food
Security, Rome, 3–5 June 2008.

Fernández P. The Equity Premium in 150 Textbooks, IESE Business School Working Paper, 2010.

Fischer G. et al. “Climate Change and Agricultural Vulnerability”, IIASA Special Report commissioned by the
UN for the World Summit on Sustainable Development, Johannesburg 2002. Laxenburg, Austria, International
Institute for Applied Systems Analysis (IIASA).

Forum for the Future, DfID and DECC Forest Investment Review, 2009, available at http://www.forumforthefuture.
org/files/Introduction_FIR.pdf.

121 Climate Change Scenarios – Implications for Strategic Asset Allocation


Goettle R. and Fawcett A. “The Structural Effects of Cap and Trade Climate Policy”, Energy Economics 31 (Special
Supplement 2): S244–S253 (2009).

Grinblatt M. and Titman S. “Mutual Fund Performance: An Analysis of Quarterly Portfolio Holdings”, The Journal of
Business, vol. 62, no. 3 (1989), pp. 393–416.

Guyatt D. and Lukomnik J. “Does Portfolio Turnover Exceed Expectations?”, Rotman International Journal of Pension
Management, vol. 3, no. 2 (2010), p. 40.

Guyatt D. “Beyond the Credit Crisis: The Role of Pension Funds in Moving to a More Sustainable Capital
Market”, Mercer’s quarterly RI Newsletter (2009), available at http://www.mercer.com/referencecontent.
htm?idContent=1332305.

Guyatt D. “A Collaborative Response to Climate Change” in Calvello A. (ed) Environmental Alpha: Institutional
Investors and Climate Change (2009).

Guyatt D. “Pension Collaboration: Strength in Numbers”, Rotman International Journal of Pension Management, vol. 1,
no. 1 (2008).

Hawker G. “Diversification: A Look at Risk Factors” (2010), available at http://www.mercer.com/referencecontent.


htm?idContent=1378620.

Holm A. “Infrastructures in the face of climate change: what implications for long term investors?”, CDC Climate
Research, Climate Report, no. 22 (2010).

Ibbotson R. and Kaplan P. “Does Asset Allocation Policy Explain 40%, 90%, or 100% of Performance?”, The Financial
Analysts Journal, January/February 2000.

International Energy Agency (IEA) World Energy Outlook 2008, Paris, IEA, 2008.

International Energy Agency (IEA) World Energy Outlook 2009. Paris, IEA, 2009.

Jordan A. and Lorenzoni I. “Is There Now a Political Climate for Policy Change? Policy and Politics after the Stern
Review”, Political Quarterly, vol. 78, no. 2 (2007), pp. 310–319.

Joskow P. and Marron D. “What Does a Megawatt Really Cost? Evidence from Utility Conservation Programs”,
Energy Journal, vol. 13, no 4 (1992), pp. 41–74.

Kristrom B. and Riera P. “Is the Income Elasticity of Environmental Improvements Less Than One?”, Environmental
and Resource Economics, vol. 7, no. 1 (1996), pp. 45–55.

Lempert R., Groves D., Popper S. and Bankes S. “A General, Analytic Method for Generating Robust Strategies and
Narrative Scenarios”, Management Science, vol. 52 (2006), pp. 514–528.

Lloyds 360o Risk Insight White Paper “Sustainable Energy Security, Strategic Risks and Opportunities for Business”
(Chatham House, 2010).

122
Ludi et al. Climate Change and Agriculture: Agricultural Trade, Markets and Investment, Overseas Development
Institute, 2007.

McKinsey and Co. Pathways to a Low-Carbon Economy: Version 2 of the Greenhouse Gas Abatement Cost Curve, McKinsey
and Company, 2009.

Mercer “Carbon Risk and Carbon Trading: Investment Considerations”, available at www.mercer.com/ri.

Meehl G. et al. “Global Climate Projections” in Solomon S. et al. (eds.) Climate Change 2007: The Physical Science
Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change.
(Cambridge and New York: Cambridge University Press, 2007)

Nakicenovic N. and Swart R. (eds.). Special Report on Emissions Scenarios, Intergovernmental Panel on Climate
Change (IPCC), 2000.

North D. “Understanding the Process of Economic Change”, London: IEA, Institute of Economic Affairs, 1999.

Parry M. et al. “Effects of Climate Change on Global Fd Production under SRES Emissions and Socio-economic
Scenarios”, Global Environmental Change, vol. 14 (2004), pp. 53–67.

Parry M. et al. Technical Summary. Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working
Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge:
Cambridge University Press, 2007.

Rahmstorf S. “A Semi-empirical Approach to Projecting Future Sea-level Rise”, Science 315 (5810) (2007), pp. 368–
370.

Rosenzweig C. and Hillel D. “Potential Impacts of Climate Change on Agriculture and Food Supply”, Consequences,
vol. 1, no. 2 (1995).

RREEF Research Globalization and Global Trends in Green Real Estate Investment (2008).

Smith P. et al. Agriculture. In B. Metz, O. R. Davidson, P. R. Bosch, R. Dave and L. A. Meyer (eds.). Climate Change 2007:
Mitigation. Contribution of Working Group III to the Fourth Assessment Report of the Intergovernmental Panel on
Climate Change. Cambridge, UK, and New York, USA: Cambridge University Press, 2007.

Solomon S., Qin D et al (eds.). Technical Summary. In Solomon S. et al (eds.). Climate Change 2007: The Physical Science
Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on
Climate Change. Cambridge, UK, Cambridge University Press, 2007.

Stern N. The Economics of Climate Change: The Stern Review, Cambridge, UK: Cambridge University Press, 2007.

Taleb N. The Black Swan, Penguin, 2007.

UNEP “Buildings and Climate Change: Summary for Decision Makers”, Sustainable Buildings and Climate Initiative
(2009).

123 Climate Change Scenarios – Implications for Strategic Asset Allocation


UNEP, SEFI and Bloomberg New Energy Finance Global Trends in Sustainable Energy Investment 2010 Report, Analysis
of Trends and Issues in the Financing of Renewable Energy and Energy Efficiency (2010).
URS Adapting Energy, Transport and Water Infrastructure to the Long-Term Impacts of Climate Change (2010).

Van der Heijden K. Scenarios, the Art of Strategic Conversation”, Chichester: John Wiley & Sons, 1996.

Vivid Economics G20 Low Carbon Competitiveness Report (2009).

Vivid Economics The Implicit Price of Carbon in the Electricity Sector of Six Major Economies (2010).

World Bank (2009a) “The Cost to Developing Countries of Adapting to Climate Change: New Methods and
Estimates” The Global Report of the Economics of Adaptation to Climate Change Study, Washington, DC: World Bank,
2009.

World Bank (2009b) World Development Report 2010: Development and Climate Change, Washington, DC: World Bank,
2009.

World Business Council for Sustainable Development Energy Efficiency in Buildings: Transforming the Market, 2009.

WWF, Trucost and Mercer “Carbon Risks of UK Equity Funds”, available at http://www.mercer.com/summary.
htm?idContent=1351415 (2009).

124
Argentina Mexico

Australia Netherlands

Austria New Zealand

Belgium Norway

Brazil Philippines

Canada Poland

Chile Portugal

China Saudi Arabia

Colombia Singapore

Czech Republic South Korea

Denmark Spain

Finland Sweden

France Switzerland For further information, please


contact your local Mercer office
Germany Taiwan or visit our website at:

Hong Kong Thailand www.mercer.com

India Turkey

Indonesia United Arab Emirates

Ireland United Kingdom

Italy United States

Japan Venezuela

Malaysia

Copyright 2011 Mercer LLC.


04028-IC-140211 All rights reserved.